Japanese Journal of Religions Studies 1992 19/2-3

Rice, Bronze, and Chieftains —An of Yayoi Ritual—

Mark J. H udson

The Yayoi 弥生 was the period in which agriculture came to form the basis of society in a laree part of the . It is often dated from 300 bc to ad 300，thoueh in parts of western wet rice farming began a century or more earlier. The end of the Yayoi was marked by the appearance in the third century of , standardized keyhole-shaped tomb mounds. Although there is disagreement over both the exact chronology of this transition and the difference between standardized and pre-standardized mounds, the majority of Japanese archaeologists now believe the Yayoi ended by about ad 250 in the west­ ern archipelago. Limitations of space make it impossible to present a general discussion of Yayoi culture here. A recent review can be found in H udson (1990), but the reader should be aware that many areas of debate still remain. One such problem is the degree of continental immigration into Yayoi Japan —something that has obvious consequences for our understand­ ing of ritual continuities with the Jomon. The Yayoi is partially proto- historic, since the Eastern Han and Wei dynastic histories (Hou H an shu 後漢書 and Wei zhi 魏志〉contain short descriptions of the 倭 people. Although there have been suggestions that “Wa” was used to refer to a distinct ethnic group, it is probably best understood as a general term for the inhabitants of at least western Japan in the third century ad. As the earliest historical account of Japan, the Wei zhi is a veritable mine of information on the customs and lifestyles of the Late Yayoi/Early Kofun periods. It tells us, for example, that “in their worship, men or import-

AC：KNOW'LEDGEMENTS ：1 wovild like to thank the foUowinsr people who were kind enousrh to provide advice and assistance of various sorts for this article: Akazawa Takeru, Gina Barnes, Choi Sung-rak, Walter Edwards, Nakayama Kiyotaka, Nakayama Seiji, Okita Masaaki, and Yamagata Manko. 140 Japanese Journal of Religious Studies 19/2-3 ance simply clap their hands instead of kneeling or bowing，” behavior that is immediately recognizable to anyone who knows modern Japan.1 Two main systems of subdividing the Yayoi are in general use. One, based on a series of five or six styles, is used primarily for western Japan; the other, a threefold division into Early, Middle, and Late phases, has a more general applicability and will be used in this paper. The Early phase began about 300 bc and lasted for around two hundred years. The Middle Yayoi occupied approximately the same amount of time and thus came to a close ca. ad 100，although recently several Kyushu archaeologists have suggested an end in the first half of that century. The Late Yayoi will be understood here as ceramic style V which was replaced by the transitional Shonai 庄内 type during the third century. Shonai was followed by Furu 布留 pottery, which marks the un ­ disputed start of the at the end of the century. This article is a review of Yayoi-period ritual practices. Rather than at­ tempting an encyclopedic coverage, I have focused the discussion around six topics. It is hoped that these topics (many of which overlap) will eive the reader some impression of the actual workings of ritual in Yayoi society. A word of warning is in order, however: I do not see these topics as defining a uniform “Yayoi religion.” The Yayoi was a period of ereat regional and chronological variation in ritual, as in everything else. At the same time，much of the cultural change responsible for this variation was negotiated through a series of distinctively Yayoi ritual structures.

R itual and the Jomon-Yayoi Transition

As the Jom on period drew to a close, ritual behavior became more and more prominent, reaching a peak in the Kameeaoka 亀ヶ岡 culture of the Tohoku region. Dating to the first millennium bc, this culture is marked by clay figurines, plaques, and masks; intricately decorated ce­ ramics with a wide range of vessel shapes; and a variety of stone objects, some of which have shapes so strange that we are at at loss what to call them in both Japanese and English. Stone circles, often with phallic cen­ terpieces, were still in existence although their popularity had waned somewhat after the Late Jomon. Most archaeologists have interpreted the increasingly ritualized nature of the Final Jomon as a reaction to deteriorating climatic conditions and to the new foreign elements that later fused to make Yayoi culture. Jom on specialist Kobayashi Tatsuo (1991，1992) has argued for a line of

1 The Wei ihi was compiled in the late 3rd century but the Hou Han Shu was not written until L50 years later and incorporated most of the earlier document’s descriptions of Japan. All translations of these Chinese histories are taken from T sunoda and Goo d rich (195 L). H u d s o n ： An Archaeology of Yayoi Ritual 141

I

F ig .1 : Sites mentioned in the te x t.(1 )Misakiyama; (2) Yatsuhagi; (3) Oki II; (4) Izuruhara; (5) Yotsuya; (6) Namani; (7) Otsuka-Saikachido; Miyanohara; (8) Miura Peninsula sites: Bishamon，Maguchi, and Ourayama; (9) Ikawazu; (10) Makimuku; Hashihaka; (11) Karako-Kagi; (12) ; (13) Tatetsuki; (14) Koura; (15) Kojindani; (16) Yoshinogari; (17) Mine; Yoshitake; Suku Okamoto; (18) Shinmachi; (19) Karakami, Iki Is. 142 Japanese Journal of Religious Studies 19/2-3

Fig. 2: Top: Bronze from Misakiyama, Yamagata. Length 26 cm. (From Kashiwakura 1961). Bottom: Stone ‘‘，，from Bussawa, . Length 35cm. (From NOMURA 1985).

Jom on “resistance” that moved slowly northwards as Yayoi culture ex­ panded; behind this line Jomon groups produced clay figurines and other traditional ritual objects in order to maintain their identity vis-d- vis the new culture. While this remains an interesting hypothesis, its details have yet to be fully developed; in many ways it is a theory that highlights the deficiencies in our understanding of Final Jomon society. At that time the archipelago was divided into two main cultural zones: the Kamegaoka in the east, and the appliqud ridge (tottaimon 突帯文) ceramic zone in the west. Although there was an area of overlap, the di­ viding line ran roughly between and the Kansai. Many readers will be aware that this is close to the traditional divide between west and east Japan, as determined from various dialectical, dietary, and cultural traits. It also marks the split between the broadleaf evergreen and deciduous forest zones. While some of the Final Jomon east-west differences may be explained by the presence of different subsistence economies within these contrasting environments, it is impossible to bring everything under that ecological umbrella. Until we understand the significance of these very different Final Jomon cultural zones, how­ ever, we cannot begin to understand the ritual transition from Jomon to Yayoi. In western Japan a trend towards plain pottery had existed from the Late phase; apart from the few sites mentioned by Kobayashi, Kamegaoka-type ritual artifacts are rare or nonexistent. Eastern Japan, in complete contrast, saw a flourishing of ritual artifacts in the Final Jom on followed by their sudden disappearance at the beginning of the Yayoi. I believe it is wrong, though, to see this ritual activity as a purely conservative reaction. Instead, new ritual patterns were both stimulated by and used to negotiate the social changes that were affecting all parts of the Japanese archipelago at that time. A distinctive type of Final H u d s o n ： An Archaeology of Yayoi Ritual 143

Jom on “stone sword/ ’ for example, may have been derived from a Chi­ nese bronze prototype. First suggested in the 1920s，this hypothesis was given further credence in 1954 with the discovery of just such a bronze knife at Misakiyama 三崎山 in Yamagata (Kashiwakura 1961). This 26- cm-long knife is thought to date from the late 2nd millennium BC. It is a rine-handled bronze knife of a type common in north in the Western Zhou. Although only one such bronze knife has been found in Japan, there is a very strong resemblance between the Misakiyama knife and the curved stone which were so important in the Final Jom on (see Nomura 1985，pp. 125-67) (fig. 2). Despite the real possibility that Final Jomon ritual was influenced by increased contact with the continent, most Jomon ritual artifacts disap­ peared abruptly at the beginning of the Yayoi. One of the best-known late Jom on ritual sites is Kinsei 金生 in .2 Both Late and Final Jom on stone pavements with uprieht stones, some worked into phallic shapes, were found with sixteen eraves. Artifacts from the site, many associated with the stone pavements, included 233 clay figurines, 560 clay earrings, nearly 30 miniature ceramic vessels, and 133 stone swords, clubs, and phalli. Many burnt deer antlers and wild

boar mandibles were also uncovered (Niitsu 8c Yamaki 1989; K id d e r 1991).The had already begun in western Japan by the time of the latest occupation at Kinsei. Recent discoveries at sites in Nirasaki on the northwest edge of the Kofu Basin mean that the earliest rice in Yamanashi can be traced back close to the time of Kinsei (see H ud so n 1991) —perhaps even within one generation. It would thus appear that the stone-based Final Jomon ceremonial facilities typified by Kinsei dis­ appeared quite abruptly around the time of the introduction of rice. From the figures given by N a k a m u r a Yoshiyuki (1989, pp. 49—50)，at least 70% of Jom on clay figurines in Hokkaido date from the Final phase; only one example is known after that. Oshima (1990，p. 83) makes the important point that the origin of many Ainu cultural traits can be found in the Epi-Jomon rather than the Jomon proper. The Epi- Jom on is the culture of Hokkaido during the Yayoi, Kofun, and early historical periods in the main islands of Japan. It was clearly a contact culture in the sense that it was formed through interaction with the ag­ ricultural, -using people to the south. Thus even in Hokkaido the beginnine of the Yayoi marks a break with what went before in many aspects of society, including ritual. O f course not all Jomon ritual artifacts completely disappeared with the Yayoi. Some, like phallic symbols and clay figurines, continued in new forms (cf. Kanaseki 8c S ah ara 1987, Plate 10; M iy ash ita 1983;

For the location of Kinsei, see the map in the preceding article by Yamagata. 144 Japanese Journal of Religious Studies 19/2-3

N iitsu 1986). Other Jomon activities, such as tooth ablation, continued until the Middle Yayoi, probably to mark rites of passage such as initia­ tion and marriage (Harunari 1987b).3 Although the idea that a society’s ritual responses are predetermined by its technological level is an over­ simplification that cannot explain the origins and development of specific behavior, we cannot deny the relation between the establish­ ment of an agricultural society in the Japanese archipelago and the more or less sudden replacement of most Jomon ritual practices by a new set of activities in the Yayoi.

The Wild and the Tame: A Ritual Menagerie

As a result of ethnographic work conducted in the 1960s, it became clear that hunting and gathering was not the nasty, brutish, and short exis­ tence that Hobbes had imagined. Indeed, Sa h l in s (1972) suggested that foraging groups formed the “Original Affluent Society.” This view had a profound effect on our understanding of the beginnings of agricul­ ture. It came to be accepted that affluent foragers would only accept ag­ riculture when forced to do so by resource stress of some sort. Farming not only involves more hours of labor but it also leaves farmers very vul­ nerable to crop failure and exploitation by social elites. Yayoi farmers would have been no less vulnerable than others. Tension was strong be­ tween the potential for the dietary and social affluence that rice could bring, and the reality of what might happen if the harvest failed, and, as we might expect, ritual negotiation of that tension was an important part of Yayoi culture. W hat is perhaps less expected, however, was the role played by wild animals, war, and hunting in that ritual. Even a quick glance at Yayoi pictorial representations on pottery and bronze shows us that animals are extremely common. Dogs’ birds, liz­ ards, frogs, and various other creatures are found, but deer seem to hold a special place in Yayoi art. In the Jomon period both deer and wild boar were major sources of food, and, judging from the bones found at ar­ chaeological sites, appear to have been exploited to a similar degree

(N i s h i m o t o 1991a). Despite this there are only two Jomon representa­ tions of deer, whereas a number of clay boar figurines exist.4 In the Yayoi the opposite situation occurs: there are many deer but very few

3 Harunari (1990，p. 28) specifically suggests the following extraction sequence: upper ca­ nines at initiation, lower canines and incisors at marriage, and upper incisors and upper and lower first premolars on the death of relatives. According to a recent newspaper report, the two Jom on sites with deer are Usvijiri 臼尻 in Hokkaido and Nishiyama 西山 in Aomori Prefecture. The latter example is a Late Jomon pottery sherd with a deer in relief. See Gekkan bunkazai sfnUsndo joho (August 1991, p. 15). H u d s o n ： An Archaeology of Yayoi Ritual 145

representations of boar (I n o u e 1990). Why is there such a difference be­ tween the two periods? It is widely argued that Jomon boar figurines were used to ensure suc­ cess in hunting (e.g., Doi 1984), but it is not known why deer were not treated in the same way. One reason may have been the fact that boar are much more dangerous and difficult to hunt than deer. An alterna­ tive explanation relates to the semi- that has been sug­ gested for Jomon boar. Neither approach really helps us to understand why deer were ignored in Jomon art, but the importance of deer in the Yayoi has been explained by a view of that animal as a local . A great deal of contextual evidence now supports this interpretation. Much of the evidence derives from pictorial representations found on and bronze bells (dotaku 銅 鐸 ) . Fifty-four pictorial dotaku are known, but as this figure includes bells cast from the same mold, the total of original works is 38. Four molds with engraved pictures have also been found, but as one of these was used to cast an extant bell, we are dealing with a total of 41 original pictorial works (H a r u n a r i 1991a, p. 443). According to Harunari's calculations, of the 364 pictorial items on these bells, deer ac­ count for 129， for 58，fish for 3 1 ,cranes for 27, but wild boar for only 18.5 Twenty-six (63.4%) of the 41 bells have deer,18 (43.9%) cranes, but only nine (22.0%) have fish and four (9.8%) boar. While the fish and boar are not always totally sec­ ondary elements，H aru­ nari points out that deer and cranes clearly form the main elements in Fig. 3: Pie diagram of animats represented dotaku art (fig. 3). on Yayoi bronze bells. “Birds” are cranes A second major source or herons; D = dragonflies; L = lizards; T of Yayoi pictorial art is = turtles; others = dog，snake，mantis， pottery. Incised drawings spider, and crab. Calculated from figures were particularly comuion given in HARUNARI (1991a).

5 H arunari (1991a, pp. 444-45) also notes the followinff possible identifications: 6 deer, 5 boar, 9 fish, and 1 . As is discussed later in the text, Harunari’s view of the dotaku birds as cranes is not shared by everyone. 146 Japanese Journal of Religious Studies 19/2-3

A j i v

Fig. 4: H a ru n a rts (1991b, p. 33) reconstruction of the picture in­ scribed on a Middle Yayoi jar from Inayoshi, . The site is also known as Sumita. on Middle Yayoi pottery in the area (Sahara 1980). Although ac­ cording to H arunari (1991a, p. 451) only 166 out of a total of less than 200 drawings are clearly intelligible, some of these are nevertheless of great importance in our understanding of Yayoi society (cf. fig. 4). Again deer are common, accounting for about 40%. A major difference be­ tween the dotaku and pottery deer, however, relates to the antlers. Only seven dotaku deer have antlers compared with 122 without; in contrast, 32 of the deer incised on pottery have antlers, 22 are without, and 30 are unclear (H a ru n a ri 1991a, Tables 1 and 3). This difference is import­ ant because Harunari argues that the annual growth of the antlers sym­ bolized the growth of rice. The eighth-century Harima no kuni fudoki 播磨国風土記 contains an account of rice germinating overnight after being planted in the fresh blood of a deer. Similar stories are known in other early texts (summa­ rized by H a r u n a r i 1991a, pp. 469-71). All these stories relate to rice- planting rituals that were conducted in the spring or early summer —exactly the time when stags lose their antlers. As noted by Kobayashi Yukio (1959, p. 5) and others, the story of Nintoku’s tomb and the account of Yamato-takeru killing a deer at Asi-gara suggest that the animal was considered a local deity. Planting rice in the deer’s blood, therefore, ensured the deity’s support in a suc­ cessful harvest. Since the dotaku deer have no antlers, Harunari infers that the bells were used in spring rice-planting ceremonies. Similarly, the presence of antlers on many deer incised on ceramics implies that those vessels were used in harvest rituals at the time of year when the antlers had regrown (H a r u n a r i 1991a, pp. 471-72). Another animal that played an important role in Yayoi ritual and be­ liefs was the bird. The fact that birds were seen as intermediaries be­ tween the secular and spiritual worlds is suggested by the Kojiki account of a funeral in which mourners dressed as birds (cf. P h i l i p p i 1968, p. 126). Wooden birds have been excavated from at least: eleven Yayoi sites. H u d s o n ： An Archaeology of Yayoi Ritual 147

Some of these have central holes by which they may have been attached to the top of wooden posts. Such a custom is known historically in and is also seen, according to some scholars, in the Wei zhi account of Mahan 馬韓 agricultural ceremonies.6 As already noted, representations of birds are common on Yayoi bronze bells. These are all long-leeged， lon^-necked birds — either cranes or herons, though it is not clear which from the actual drawings. H arunari (1987a，pp. 17—18) areues that they are cranes from their body stance, but Terasawa (1991a, p. 170) and others have made the obvious point that since cranes migrate to Japan in the late autumn, such an identification might imply a quite different ritual role from that of herons, which arrive in the summer. A myth found in the texts and in later historical Japan attributes the origins of farming to a bird that brought down rice seed in its mouth. H a r u n a r i (1987a，pp. 21-22) believes that a similar belief existed in the Yayoi and that an annual ceremony to call back the rice spirit symbol­ ized by the bird was held to ensure a plentiful harvest. As “the world of the birds was across the sea, the priests donned bird outfits, boarded a boat, and went out to meet [or call] the birds” (H a ru n a ri 1987a, p. 38). Such scenes, which recall the Kojiki story of Sukuna-biko-na (cf. Philippi 1968，p . 115)，are known from a number of pottery vessels. Fie. 4 shows the incised drawing on a Middle Yayoi jar from Inayoshi 稲吉 in Tottori Prefecture. Harunari interprets the oarsmen as feathered “priests” row­ ing towards a raised-floor storehouse; Kanasek] (1986，p. 295) sees the objects to the left of this as two bronze bells hanging from a tree, based again on the Wei zhi account of a Mahan ritual facility with bells and drums suspended from a tree. A deer also forms part of the design’ al­ though its original position on the vessel is not dear. Drawings of both boats and priests with feather-like plumes are known from a number of other ceramic vessels, many of which are illustrated in Harunari’s rele­ vant articles. Many of the above aspects of Yayoi agricultural ritual come together in the scenes depicted on a bronze thought to have been found in Takasaki, Gunma Prefecture, and now in the National Museum. This is despite the fact that the mirror itself is believed to date to the Early Kofun period. As shown in fie. 5, the decoration consists of two concentric bands. In the inner band four figures are interspersed with four deer. Two of the figures carry long poles, one also has a bow, and one is holding a ja r aloft. They are strange, faceless figures with three- pointed webbed feet and hands. O f the deer, two have antlers, one does not, and the fourth is unclear since the mirror is damaged in that place.

6 Mahan was the name given to a region of southwest Korea in the Proto-Three Kingdoms period (cf. Barnes 1990，pp. 131-35). The presence of bird poles in the Wei zhi account of this region is based upon interpretations of the word stUu 蘇塗(Kanaseiu 1986, pp. 287-92). 148 Japanese Journal o f ReLigiom Studies 19/2-3

The outer band has ten figures, nine of which are armed with shields and swords or . The one unarmed figure stands with arms raised per­ pendicular to the circumference of the mirror. Most of the outer figures have frond-like extensions from their bodies that recall the priests' “feathers” on the Inayoshi jar. O n the basis of

^ C 1 ° Fig. 5: from Gunma Prefecture now the deer and the in the Tokyo N atio n a l Museum. From S iiit a r a pottery jar (which . is believed to have been used to store rice grain for sowing in the following season), S hitara (1991a, p. 60) interprets the inner band of the mirror as depicting an ag­ ricultural ritual. The Wei zhi account of the Mahan mentions both spring and autumn farming rituals; Shitara argues that both seasons are rep­ resented here also, since there are deer with and without antlers, and since the figure with the jar could signify either season, sowing seed in the spring or collecting it in the fall. Shitara compares the outer band of this Gunm a mirror with the Mahan ritual dancing described in the Wei history; he also cites N a k a m u r a Tomohiro’s (1987) suggestion that the wooden weapons found at a number of Yayoi sites were used in mock battles. S h it a r a (1991a, p. 6 1 )thus hypothesizes that the outer band of the mirror represents such mock fighting as part of an agricultural ritual. We have seen that the contextual evidence supporting the role of deer and birds in Yayoi ritual is very good, explaining to some extent many Yayoi ritual artifacts and their designs. It. seems to me, however, that we should and can go further. In particular, Japanese archaeologists have yet to consider the question of how these ritual practices are related to the overall structures of Yayoi society. Why were deer and birds used in agricultural ritual? What was the relationship between rice farming and the hunting and war scenes found on r.hc ddtaku and the Gunma mirror? In a recent book, Ian H o d d e r (1990) has argued that the domestication H u d s o n ： An Archaeology of Yayoi Ritual 149 of plants and animals the domestication of society in terms of the social control and constraints needed to build common facilities and structures. In Europe this social domestication was negotiated in various ritual ways, often involving the tension between the wild and the tame. Interestingly, similar patterns can be seen in the Yayoi. In con­ cluding this section I would like to suggest that there was a link between Yayoi agriculture, warfare, and hunting, a link that provided both a met­ aphor and a rationale for the control of society. The Yayoi period is thought to have been an age of “war and weap­ ons” (Sa h a r a 1987b，pp. 283-98; and in press). This interpretation is based on the presence of true weapons, defensive settlements, remains of burnt houses, skeletons with embedded sword and points, and the Wei zhi account of the “Wa Unrest.” In such a society there was a need for heightened social cohesion to construct moats, watchtowers， and so on —and to fight if necessary. There were thus ready opportuni­ ties for individuals or lineages to gain power within Yayoi society. The need for warfare and social cohesion was negotiated through the ritualization of war and its association with hunting. As will be discussed later, bronze daeeers, spearheads, and halberds all evolved from actual weapons at the start of the Yayoi to wide-bladed, impractical ritual ob­ jects. Weapons, in other words, became objects of worship. Ritual weap­ ons were also made of wood and were probably used in performances such as those described in the Wei zhi and depicted on the Gunma mir­ ror. This mirror provides a clear link between (mock) warfare and the hunting of deer. If we agree that the deer represented land spirits in ag­ ricultural ritual, then I believe we can understand why hunting and warfare scenes are so common in Yayoi art. Hunting and warfare were a metaphor for the control of the wild or, in other words, the domesti­ cation of both rice and society. This relationship may be illustrated as follows:

Hunting (= control of) deer: agriculture (ニ control of) rice: warfare (~ control of) society

This relationship is complicated but not necessarily invalidated by the presence of domesticated pigs in the Yayoi. Though suspected for some time, the existence of pies was confirmed only in 1989, when - preserved skulls were found at the Shimogori-kuwanae 下郡桑苗 site in

Oita (N i s h i m o t o 1989). These animals are now known from at least eieht sites in western Ja p a n / Many examples display evidence of the

7 Since it is difficult to separate wild boar from pig unless the bones are especially well- preserved, this implies pigs were quite common in the western Yayoi. 150 Japanese Journal of Religious Studies 19/2-3

ritual treatment of jawbones whereby a hole was cut at the back of the lower jaw and the bones were then suspended on a wooden pole. Pig jawbones still attached to poles were found at Nabatake 菜畑 in Saga and Karako 唐古 in Nara.8 Since this ritual is known from the very beginning of the Yayoi, it is thought to have been brought to Japan by continental immigrants (Nishimoto 1991b). The exact nature of the jawbone cere­ mony is still poorly understood, however, and in any event it was appar­ ently limited to the Yayoi. Somewhat mysteriously, domesticated pigs seem to have disappeared completely from Japan after the (see S ah ara in press).

Oracle Bones and Divination

Divination has played an important role in Japanese religion since very early times (Blacker 1981，1983). The term refers to com m unication with the supernatural world for the purpose of determining an appro­ priate course of future action through the answering' of questions that transcend human understanding. Divination by the application of heat to animal bones was found widely throughout Eurasia. It reached a peak in Shane-dynasty China, when cattle and water bufTalo scapulae and turtle shells were used, many being inscribed with the result of the div­ ination. The first direct evidence we have of divination in Japan is the oracle bones of the Yayoi period. Divinatory practices are also men­ tioned in the Wei zhi:

Whenever they undertake an enterprise and discussion arises, they bake bones and divine in order to tell whether fortune will be good or bad. t irst they announce the object of divination, using the same manner of speech as in tortoise shell divination; then they examine the cracks made by the fire and tell what is to come to pass.

Over 100 oracle bones have come from around 25 Yayoi-period sites. The exact figure varies between authors: Kanzaw a(1990，p. 8 2 )lists 116 examples, whereas H a r u n a r i (1991a，p. 466) has only 101, There is also disagreement over the date of the earliest bones. An example found at the Koura 古浦 site in is included by H a r u n a r i (1991a, p. 466) but discounted by Kanzawa (1990，pp. 84-85) on the grounds that it was not clearly associated with datable artifacts and that the style of burning suggests a Kofun-period date. Oracle bones from

8 The Nabatake find is illustrated in H udson and Barnes (L991 p. 232). Kofim-period rit­ uals using wild boar bones are mentioned by Ishino in this issue. Kanzawa’s figures in his Table 2 do not match, but elsewhere in the text and in his Table 5 (Table 1 of this paper) he mentions a total of 116. H u d s o n ： An Archaeology of Yayoi Ritual 151

Karako, are assigned to the Middle and Late phases by Kanzawa, but the excavators of the site clearly put some of them in the Early Yayoi (F u jit a 1986, p. 4 7 , 1988，1989). The Karako examples would thus appear to be the earliest oracle bones in Japan; all others date to the Middle Yayoi or later. While to some extent biased by preservation factors, the distribution of oracle bones shows that divination was common throughout much of Yayoi Japan except for Kyushu and the Tohoku. Oracle bones are especially common in the Kinai and south Kanto. So far none have been discovered from Kyushu proper— only Iki and (in the Kofun period) Tsushima Is­ lands. By far the majority of Yayoi oracle bones have come from coastal sites, but this seems to be related to preservation as much as anything else. Bones are better preserved in the alkaline conditions of shell mounds, and such contexts account for 69% of the sites. Finds at inland locations such as Yotsuya 四ッ屋 and Namani 生仁 in , however, show that divination was not performed only by coastal populations. Table 1 lists Yayoi oracle bones by species and body part Deer scapu­ lae were by far the most commonly used bones. The choice of deer was probably related to the ritual importance of that animal, as discussed in the previous section. It should be noted that divination usine turtle plas­ trons seems not to have appeared in Japan until the Kofun era.

Deer Wild boar Deer Dolphin Species TOTAL (Cervus nippon) (Sus scrofa) or boar (Delphnuclae) Indef. Scapula Rib Verte- Meta­ Scap. Scap. Rib Indef Vertebrae Bone brae tarsal Indef. 7115 11 16 3 5 2 1 1 116

88 16 7 1 1 116 75.86% 13.79% 6.03% 0.86% 0.86% 100%

Table 1 .Yayoi oracle bones by species and body part. From Kanzawa 1990，p. 87.

Almost nothing is known about the actual divination rites of the Yayoi. Examination of excavated oracle bones has shown that the surface of the bone was polished and sometimes sliehtly shaved. Heat was then applied to points on either side, causing cracks to appear on both sides of the bone. According to Ka n za w a (1990，p. 96), all Yayoi oracle bones used this technique, which is not found in other periods. We are completely ignorant as to how the cracks were interpreted and the aim of the divi­ nation. Ka n z a w a (1990, p. 103) suggests that divination was mainly con­ ducted for matters concerning the agricultural cycle and warfare, and this hypothesis may not be too far off the mark. Another problem is knowing by whom the divinatory rituals were performed. In later 152 Japanese Journal of Religious Studies 19/2-3

Fig. 6: Yayoi oracle bones:(1)Karako-Kagi, Early Y ay o i.(2) Bishamon Cave C，Late Yayoi. (3) Maguchi Cave, Late Yayoi. (4) Karakami, Iki Is­ land, Late Yayoi. From FUJITA (1986) and Kanzawa (1990). periods there were specialist hereditary groups of diviners known as urabe 卜部 (cf. Hashiguchi n.d.)，but divinations at the family or village level seem more likely for the Yayoi (Kanzawa 1990, p. 104). Oracle bones were not found in the Jomon and were almost certainly introduced from the continent in the Yayoi, The direct source was prob­ ably Korea, where at least five sites have produced oracle bones (Wata- nabe 1991). One of these sites —Hogok 虎谷 in North Korea — may date to a few centuries earlier, but the others are all from about the same time frame as the Yayoi. Most —but not all —peninsular oracle bones appear to have been subjected to the same divination technique as the Yayoi ex­ amples, and after use they were disposed ot in a similar fashion (Wata- nabe 1991，p. 5). Deer scapulae were also common in Korea. At the Kunkok-ri 郡谷里 shell mound at the southwest tip of the peninsula, deer replaced boar bones as the favored material after the 1st century bc (Watanabe 1991，p. 8). Although the Korean evidence is currently less abundant than the Japanese, the presence there of oracle bones and a few deer representations on bronzes suggests that the deer may also have played an important ritual role on the peninsula. Thus, as H a r u ­ n a r i (1991a，p. 475) points out, there is a possibility that Yayoi beliefs about the deer as a local deity in agricultural rituals may have been transmitted with oracle-bone divination from Korea. H u d s o n ： An Archaeology of Yayoi Ritual 153

Bells and Blades: Bronzes, Hoards, and R itual Zones

The Yayoi period saw the introduction of both bronze and iron in 10 Japan. Utilitarian objects were made of iron, as well as wood and stone; bronze, in contrast, was used almost entirely for ritual implements. The two most characteristic Yayoi bronze implements were bells (ddtaku) and weapons. Both of these were initially imported from the continent, but soon thereafter were produced in Japan. Both also underwent a similar process of exaggerated enlargement as they changed from practical ob­ jects into ritual and political symbols. In many ways ddtaku are an archaeologist’s nightmare, because almost, all have been found without any associated artifacts that might give clues to their date or function. They probably derived from a type of small bronze bell known on the Korean peninsula and also found in Kyushu (Sahara 1987a, pp. 49-54), but there is still controversy over ex­ actly when and where they were first produced in Japan. Sa h a r a (1960) worked out a typological sequence based on the handle. Early triangular handles becamc rounder, wider, and more ornate as their original func­ tion of suspending the bell was lost. Sahara’s Style II often has a flowing- water pattern common on Kinai ceramics of the early Middle Yayoi; assuming bell and pottery decorations were more or less contemporary, Sahara suggests the earliest were made in the Kinai from the end

of the Early Yayoi, a conclusion supported by K u r a k u (1990, p. 44). Such a date was oncc seen as anomalous, sincc it appeared to preccdc the first bronze casting in Kyusliu, contradicting the traditional view among Japanese archaeologists thai bronze casting spread from west to cast along the archipelago. Although more scholars now accept the pos­ sibility of parallel origins in both Kyushu and the Kinai (e.g., Kondo 1987a，p . 116)，rcccnt finds at Yoshinogari 吉野ヶ里 in Saga have put the beginnings of metalworking in Japan well back in the Early Yayoi, al­ though it must be stressed that no such secure cvidencc has yet come from the Kinai. In addition, some archaeologists such as Terasawa (1991a, pp. 149-50) arouc on typological grounds that dotaku produc­ tion beean in Kyushu and spread east. As ofjanuary 1992 there were 430 dotxiku known from 302 sites. Fif­ teen bell fragments and the 62 dotaku whose provcnancc is unknown should bc added to these to eive a grand total of 507. At present, the whereabouts of 112 of l.hcse bells arc unknown (Okita in press). Most of the bells were found buried on isolaied hillsides away from settlements, and there can be no doubt that they were buried there intentionally. To explain such behavior is difficult, but a number of possibilities exist:(1) underground storage as part of an annual ritual cycle; (2) concealment in times of unrest; (3) abandonment; (4) some sort of votive offering. Sa - h a r a (1960; 1987b, pp. 280-81) has arirucd for the first of these theories, 154 Japanese Journal of Religious Studies 19/2 -3 suggesting that dotaku were kept buried in the soil in normal times and only brought out for special festivals. The ethnologist M ish in a Shoei (1968) proposed that the buried bell became a receptacle for the earth and crop spirits, which were then brought back from the soil when the bell was exhumed. Mishina thought that the disappearance of dotaku at the end of the Yayoi represented the transition from such an earth-cult to a heaven-based one in the K ofun era, a transformation for which sup­ port can be found in the Kojiki. Terasawa (1989，p .フ1 )doubts that bells were regularly buried and ex­ humed, since (in the cases where such information is available) there is never more than one pit at a hoard site and it seems to have been sealed with clay or fine soil. He also points out that no attempt appears to have been made to protect the bell in its pit, yet its functions both as a musical instrument and as a symbol of power would presumably have been im­ paired by letting it rust underground for part of the year. Harunari (1987a，p. 4) has criticized Mishina’s earth-cult concept on the grounds that it is extremely difficult to match evidence from mythological and ar­ chaeological sources. Mishina himself acknowledged that the earth cult was still present in the fudoki of the Nara period and thus any change must have been gradual. According to Harunari, the thick borders of dotaku designs symbolize its role of confining the rice spirit:

The bronze bell was a magical vessel that confined grain spirits within the unhulled rice, that guarded them from other earthly spirits such as drought, heavy rain，and wind. Therefore, the place where the bronze bell was stored was the same place where grain spirits were stored. W hen grain spirits dwelled in the rice paddies after the sowiner [and] before the harvest，the bells m ight have been kept in the storehouse and taken out to riner when necessary. (H a ru n ari 1982, p. 330) While this scenario may seem attractive, it does not rule out burial over the winter when the grain spirit returned to its home. As Terasawa pointed out, however, the context of most bell finds implies a final rest- inp-place. The question of the final burial of ddtaku is inseparable from the problem of why the bells dropped out of use at the end of the Late Yayoi. An explanation seems to be linked with their changing function against a background of increasing social complexity. As dotaku are hardly ever found in houses or , Kobayashi Yukio (1959，pp. 50- 5 1 )argued that they were used and possessed by communities rather than individuals or powerful families. We have already mentioned Kanaseki’s interpretation of the incised pottery sherd from Inayoshi as two dotaku hanging from a tree (fig. 4). Okita has argued that two bells were standard for each ritual unit, an idea that is supported by the fact H u d s o n ： An Archaeology of Yayoi Ritual 155 that the majority of finds have been of even numbers (O kita m press, and personal communication). Thus hoards with more than two ddtaku — which became common in the Late phase — may have been made by larger regional units than previously existed. Also in the Late Yayoi, ddtaku became bigger and more ornate but no longer depicted agricul­ tural ceremonies. Tanaka (1970) has called these “bells for looking at” in contrast to the earlier “bells for listening to•” H a r u n a r i (1991b, p. 38) suggests that a similar degeneration can be seen in the change from in­ cised drawings to abstract signs on pottery, and that the “bells for listen­ ing to” were buried at this stage. We cannot really understand bell hoards without discussing bronze weapons, which were deposited in a similar fashion. There were three main types of Yayoi bronze weapons: swords, halberds，and spearheads. Local castine had begun by the end of the Early phase, and weapons produced in Japan developed quickly from their original forms into blunt, wide-bladed, “weapon-shaped ritual objects.” J list as “bells for lis­ tening to” became “bells for looking at,” so “weapons Lo kill with” be­ came ceremonial “weapons to look at” （S ah ara 1987a, p. 50). Since the end of the last century it has been noted that bronze weap­ ons were found mainly in north Kyushu, whereas bronze bells were cen­ tered in the Kinai. From an early stage historians and archaeologists have argued that these represented political and/or ritual zones (see Tanaka 1991，pp. 167-74). Particularly influential was the suggestion in 1939 of the philosopher W atsuji Tetsuro (1977) that the Kyushu polity represented by the “ zone” moved east, conquering the Kinai ddtaku polity and forming the Yamato state; Watsuji linked this with Lhe eastward expansion of Emperor Jinm u as recounted in the Nihon shoki. The reality, of course, is not as straiehtforward as Watsuji supposed. To begin with, the bell and blade zones were never totally exclusive. We have already mentioned the possibility that bells were originally pro­ duced in Kyushu. Although actual ddtaku have not yet been found there, eight casting molds are known from Saga and , and bells may have been made in Kyushu and transported to the Chueoku region (H a ru n a r i 1989). Similarly, ritual bronze weapons are also known in the Kinai and surrounding regions. A number of hoards are known where different types of bronze implements were buried together. The most spectacular example of this is Kojindani 荒神谷 in Shimane, where, between 1984 and 1985，six bells and 16 spearheads were found buried together some 7 m from another pk containing 358 bronze swords. As U eda (1986, p. 403) pointed out, this unprececlenLecl discovery “has stiinulaLecl a reexamination of the peculiarities of the bronze cultur[al] sphere[s] of the time, as well as a rethinking of the likely political configuration of the ancient Izuino area.” The question of chronological variation is crucial in understanding 156 Japanese Journal of Religious Studies 19/2-3

Yayoi bronze distributions. Details vary between individuals, but most archaeologists agree that bronze zones became more polarized over time. Three major stages can be recognized. In the first，bell and weapon production began in Kyushu or perhaps simultaneously in Kyushu and the Kinai. Either way, regional interaction was high, culmi­ nating in the second stage in the latter half of the Middle Yayoi. This was when the zones overlapped the most (cf. T erasawa 1991a，p. 177). In contrast, the Late Yayoi saw the most polarization: the zones posited by T erasawa (1991 a, p. 179) and K o n d o (1987b, p. 175) hardly overlap at all, although a few stray finds still come from outside their respective territories. Fig. 7 shows Rondo's six Late Yayoi ritual zones. Terasawa (1991a, p. 181) argues that the two Chugoku sword zones may not be contempo­ rary and he omits them from his scheme. A more serious problem is the social meaning of these distributions. Japanese scholars have usually as­ sumed they represent actual sociopolitical units of some sort. As Barnes (1988，p . 12) reminds us, however, “it is an axiom in anthropological re­ search that the distribution of cultural traits does not unequivocally imply political integration. . . Thus we cannot automatically assume that the bronze zones directly reflect separate ethnic or political units. At the same time，we cannot ignore the possibility that bells and weapons were used as symbols to cement regional alliances, just as keyhole­ shaped tombs served a similar purpose in the Kofun period. This sce­ nario is one favored by most Japanese archaeologists, who see the fixed bronze zones of the Late Yayoi as reflecting increasing political unification. In particular, the ddtaku/h2L\berd division is thought to rep­ resent growing tension between large chiefdoms based in the Kinai and north KyQsliQ (H a r u n a r i 1989; K o n d o 1987c，p. 104; Terasawa L99Ia). Over the 400 years of their use, Yayoi bronze implements underwent major changes from original secular bells and weapons to ritual objects used in agricultural ceremonies of the type discussed in the previous section. In the Late Yayoi they took on an increasingly symbolic power through association with political elites. They were finally abandoned at the end of the Yayoi when those elites had become so powerful that they were no longer in need of the ritual legitimation provided by the tradi­ tional bronzes. Power structures were now negotiated through a new rit­ ual system based arouncl mirrors and tomb mounds. That system will be discussed later in this article.

Life after Death

An almost bewildering variety of burial types appears in the Yayoi. Many were short-lived or confined to certain regions，while others spread

H u d s o n : An Archaeology of Yayoi ritual

Late Yayoi ritual zones proposed by KONDO (1987b). Hondo's zones and5 6 are thought to ： Fig. 7 be of different date by some other scholars: zone 5 is medium-narrow is 5 differentbronzebronzedaggerszoneandotherofscholars: zone6 some bedate by daggerswith sideprojections. 158 Japanese Journal of Religious Studies 19/2-3 across wide areas or persisted for many centuries, with some continuing into the following Kofun period. A number of burial types were discov­ ered only in recent decades and are thus not mentioned in earlier En­ glish reviews such as Ka n ek o (1968). The Yayoi period can be said to mark the beginning of large-scale formal in Japan. Unlike in the Jomon, when many burials were made inside the village perimeter, Yayoi graves were usually separated from the actual settlement area. Very few Yayoi burial types seem to have been directly copied from the continent, but one exception was the megalithic graves of Kyushu, which were clearly inspired by Korean . In this type of burial, the corpse was placed in an earthen pit, a jar, or a wooden or stone coffin; the was then marked by a monolithic capstone (fig. 8，no. 4). Though transmitted from Korea as early as the Initial Yayoi, these megalithic graves had a limited geographical and chronological distri­ bution in Japan, centered in Saga and Nagasaki prefectures in the Initial and Early phases (see I wasaki 1987).10 Contrary to expectation, this burial custom does not seem to have been limited to continental im ­ migrants. This is clear from the distribution of the , from the , and from the discovery of skeletal remains with Jomon characteristics from Initial and Early Yayoi megalithic graves at the Shinm achi 新町 site in Fukuoka (N akahashi and N agai 1987). N ot only were megaliths much less common in Japan than in Korea, they also seem to have been least popular with populations that came from the peninsula. An explanation for this unusual situation has yet to be found. In the Fukuoka Plain area, pits with or without wooden coflins were the main burial type of the Early Yayoi. By the end of the Early phase burial jars for adults had appeared, niarkine the start of one of the most distinctive Yayoi burial customs 8, n o . 1 ).These jars were made specifically for use in primary burials. They should thus be distin­ guished from the small jars used for the of adults known in eastern Honshu. Primary adult jar burials were not found out­ side north Kyushu, except for a few examples in the south of that island. Although their appearance corresponds with a large influx of bronze, pottery, and other influences from the peninsula, adult primary jar burials are also unknown in Korea at this time and seem to have been an indigenous development. An early site is Yoshitake-Taka«*i 吉武高木 in Fukuoka City, where a dating from the end of the Early to the beginning of the Middle Yayoi contained 34 jar burials (of which 16 were for adults) and four wooden coffins. All of the coffin burials and ei^ht of the jar

10 The term “Initial Yayoi” is used by some Japanese archeoloeists to refer to the period from the first evidence of rice aericiilture to the beginning of the Early Yayoi (see H udson 1990，p. 65). H u d s o n ： An Archaeology of Yayoi Ritual 159

Fig. 8: Main Yayoi burial types.(1 )Primary jar burial from Kaneno- kuma, Fukuoka. Total length about 154 cm; (2) Secondary jar burial pit at Izuruhara, Tochigi. (3) Anthropomorphic ceramic container from Oka, Yamanashi, 22 cm; (4) Megalithic burial, Shinmachi, Fukuoka. Length of capstone is 180 cm; (5) Square moated burial precinct from Nojima, . About 12 x 12 m; (6) Mound burial Y-l at Kami, ; (7) Four-cornered mound from Chusen-ji, Shimane. Length, ca. 30 m. 160 Japanese Journal o f Religious Studies 19/2-3

Plate 1 :Grave goods from coffin No. 3 at Yoshitake-l^kagi, Fukuoka. Properly of -cho; reproduced by permission. burials included grave goods. Most were limited to jasper, jade, or glass beads and/or a single bronze dagger, but in coffin no. 3 a bronze halberd, spearhead, and geometric-incised mirror were found in addition to two daggers (see Fu k u o k a 1986). This grave, dating to the beginning of the Middle Yayoi, is thus the earliest example of the mirror-sword-jewel combination that was later adopted as the imperial insignia (Plate 1). Jar burials se cannot be interpreted as elite graves, since many sites have produced hundreds or even thousands of such jars, the majority of which contain no artifacts of any sort. Some jars, however, contain splendid collections of bronze weapons, mirrors, and other prestige goods. One jar at the Suku Okamoto 須砍岡本 site produced around 30 imported mirrors or fragments thereof. Such rich graves have been in­ terpreted as the burials of the chieftains of the polities mentioned in the Wei zhi. These rich graves were often demarcated from other jar burials by position or by placement in an actual mound. Traditionally it was thought that burial mounds only appeared in Japan in the Kofun period, but over the past twenty years or so it has become clear that mounds also existed in the Yayoi. There are a wide variety of Yayoi mounds and as many different opinions as to how they should be classified. As might be expected, terminology is a nightmare. The researchers can be roughly divided into two camps: the lumpers and the splitters. My own feeling is that lumping any grave with a raised mound into a single category may mask potentially meaningful varia- H u d s o n ： An Archaeology of Yayoi Ritual 161 don that is not yet undersood. I thus follow the terminology used in Kanaseki and Sa h a ra (1987)，although I realize that it is not always easy to clearly distinguish between the categories they define. The earliest Yayoi grave that can be called a mound-burial was exca­ vated at the Mine 峯 site in Fukuoka. Dating to the first half of the Early Yayoi, it is a rectangular some 18 m lo n g ,13 m w ide,1 m high, and surrounded on three sides by a 1.7-m-deep ditch. This ditch may have continued on the fourth (south) side, but it had already been de­ stroyed prior to excavation. At least seven pits that had possibly con­ tained wooden coffins were found within this “mound” （Kataoka 1991， p. 190; O d a 1990，pp. 91-92). The trend to demarcated burial features separate from other ordinary graves is clearly seen at Yoshinogari in Sapa，where an oval mound some 40 m long and 30 m wide dates from the early first century BC. Its present height is 2.5 m, but it may origi­ nally have been twice that size. Within the mound, the actual burials were placed in large jars. In that respect they did not differ from the jar burials of the contemporary cemetery surrounding the mound. In terms of grave goods, and placement within the huge mound, however, the in­ dividuals entombed there were no doubL of the highest status at Yoshinogari (Saga Board of Education 1990，pp. 67-75; Hudson and Barnes 1991，pp. 219-24). Saving further consideration of Yayoi mounds until later, we must now look at eastern Honshu, where quite different burial practices ex­ isted from those in the west of Japan. In the Early and first half of the Middle Yayoi, secondary burial was widely practiced from the Tokai to the south Tohoku regions. This custom was then replaced by moated burial precincts, which spread from the Kinai area. These precincts were only found as far north as Niigata and Ibaraei; beyond this line pri­ mary pit burials, often containing jars, were the norm. Secondary burials (saisobo 再葬墓) involved the placement of human bones inside ceramic jars that are too small to have contained a primary adult burial. In the Yayoi they are thought to have appeared simulta­ neously in the Mikawa and central Fukushima regions from the middle of the Early Yayoi: burials with one jar pit spread from the former re­ gion and pits with multiple jars from the latter region (Shitara 1988). There is considerable controversy over the origins of Yayoi secondary burials. S u g ih a r a (1968) and H o s h id a (1976) looked to the jar burials of the Final Jomon, but Ishikaw a (1987，p. 152) notes that almost all such burials in eastern Japan are the primary interment of children. Ishikawa (1981，1987)，stressing the process of secondary burial, pro­ poses Final Jom on bone pits and the reburial of cremated remains as more likely ancestors to this custom. Through recent work associated with the re-excavation of the Ikavvazu 伊川津 site in Aichi, it has become clear thal secondary burials did in fad exist in the Final Jomon 162 Japanese Journal of Religious Studies 19/2-3

(H a ru n a r i 1988). Instead of ceramic jars, bones were reburied in simple pits. Cremated bones have also been found from this phase, providing another link with Yayoi practices (H ar u n a ri 1988, pp. 413-16; I s h i­ kawa 1988). Our understanding of Yayoi secondary burial has been greatly ad­ vanced by the work of a group of Gunma archaeologists (Iijima et al. 1986，1987; Aram aki et al. 1988; Toyama et al. 1989). From a detailed analysis of the bones and teeth excavated from the Early Yayoi Oki 沖 II site in Fujioka City, Gunma, they were able to suggest the following se­ quence (Aramaki et al. 1988，pp. 81，98):(1)death; (2) burial of the corpse in an earthen pit; (3) decomposition of soft tissue; (4) exhuma­ tion of bones; (5) removal of several teeth for wearing by relatives; (6) insertion or bones and remaining teeth into a ceramic jar that was then buried in a separate pit; (7) of bones that could not fit into the jar, followed by their return to the main pit; burnt animal bones were also included at this stage, possibly as offerings; and finally, (8) re­ turn of teeth worn by relatives to the main pit after their deaths. Al­ though it has not been possible to reconstruct al] of these stages at any other site, at least some are present al many. Two reports on actual burial-jar pit sites excavated by University have quite long English summaries (Su g ih a r a and O ts u k a 1974; S u g ih a r a 1981). Cremated bones came from the Yatsuhaei 八束脛 cave in Gunma and human bones, and teeth with holes pierced for use as ornaments, are known from several sites (Miyazaki et al. 1985). Cut marks on bones from the Ourayama 大浦山 and Maguchi 間口 on the Miura Peninsula are interpreted by H a r u n a r i (1988，p. 417; 1991c，p. 101) as evidence of defleshing for secondary burial rather than cannibalism (an earlier in­ terpretation by Suzuki [1983，pp. 161-66]). For various reasons interpretation of the meaning of secondary burial in Yayoi society is difficult. The actual bones were disarticulated and/or cremated so that they fitted into jars with narrow necks often of about 10 cm diameter. As these jars are usually found separate from settlement remains，the connection between village and cemetery is generally un­ clear. T anaka (1991，p . 115) writes that jars were buried in their pit on a single occasion.11 He further notes that many burial jars bear traces of having been carefully repaired, something which has been remarked upon by others. O f course，by its very definition multiple secondary burial involves a situation where individuals die at different times but are eventually reburied together. If Tanaka is correct, however, reburial may have been conducted only on special, limited occasions. As a possible ex­

11 I have not seen this observation made anywhere else and suspect it is difficult to prove for most early excavations. H udson ： An Archaeology of Yayoi Ritual 163 planation for this Tanaka proposes the existence of age groups, mem­ bers of which were buried together. This hypothesis would be better supported if individuals of the same sex and similar age were found bur­ ied together at a significant rate. Since no such examples have been found, Tanaka’s hypothesis remains without supporting evidence. Al­ though he further suggests that this type of age group-centered society was a continuation from the Jomon, this would seem to be ruled out in the case of grave no. 6 at Ikawazu, which contained two mature males, eight mature females, and three infants. A related problem is the degree of social stratification represented by these secondary burials. Tanaka (1991, p . 115) takes the view that there were no status differences in east Honshu at this time, since everyone was buried in similar jars under similar conditions. Others, though, have interpreted the presence of grave goods and jars with sculptured faces as evidence of stratification. Su g ih a ra (1981，p . 15) saw the beads from the Izuruhara secondary burials as evidence for the existence of upow- erful chieftains，” though it should be noted that such grave goods are rare at most other sites. Shitara (1991b，p. 199) writes that since there is never more than one jar with a sculptured face in any given site, that jar must have contained the remains of a person with special standing. W ith secondary burials, however, the link between grave goods and status is not direct. As T illo t so n (1989, p. 7) points out，“A simple pot of secondarily transferred bones usually implies considerable funerary expenditure, even without other material remains•” Secondary burial is a long term process with much scope for reinterpretation of status after death. For this reason it can be seen as a transitional custom, preceding the later moated burial precincts where status was already fixed before death. The moated precinct was one of the most common types of Yayoi burial. They are normally either square or circular and contain wooden-coffin or pit burials in the center of the precinct and sometimes in the moat. Some Japanese archaeologists include these under the category of mound-burials. At the very least, the soil dug from the moat is likely to have been piled up inside the precinct. In most excavated examples, however, little or nothing of the original mound remains. More import­ ant than the mere height of the burial is the role played by the grave in Yayoi society. Mound-burials such as thatat Yoshinoeari were clearly quali­ tatively different facilities, designed for use by only the most powerful in­ dividuals or families. Despite regional and temporal variation, moated precincts are interpreted as the resting ground of a wider segment of society. The first moated burial precincts appeared in the Kinai around the middle of the Early Yayoi. From there they spread to the Kanto by the Middle phase and to Kyushu from the end of the Yayoi to the beginning of the Kofun period (Ta s h ir o 1987). W ithin the basic form of a central 164 Japanese Journal of Religious Studies 19/2-3 precinct surrounded by ditches, there is considerable variation in the way the moat was divided or “bridged.” The division of space repre­ sented by these burial precincts can be seen as a reflection of settlement space at a time when many villages were also surrounded by deep moats (Mizuno 1990，pp. 97-104). The ethnologist Obayashi Taryo (1987, pp. 170-72) has suggested that a symbolic link between the central burial area and the Onogoro island of the Kojiki lancl-creation myth might have served as a symbol of rebirth. It is not clear, however, how burials in the moat of the precinct might fit into such a scheme. In the Kanto, moated burial precincts first appear in the Suwada 須和田 phase of the early Middle Yayoi. This stage marked a major turn- in? point in that region since it witnessed, along with new mortuary cus­ toms, the spread of moated villages and full-scale wet-rice agriculture. These trends culminated in the following Miyanodai 宮ノ台 phase, when there was an explosive increase in the number of sites. The switch from secondary jar burials to moated precincts may be said to represent a major shift in beliefs about treatment of the dead and in social complex­ ity, though it is possible that the same basic type of social organization was behind both burial practices. This is because the corporate groups stressing ancestral rights to land thought to be represented by moated precincts would also be a good way to explain secondary burials. In contrast to the lack of consensus on secondary burials, it is quite widely accepted that these precincts were the graves of some sort of kin­ ship-based agricultural production unit. Generally speaking, at any one site moated precincts are too numerous to have been elite burials, but too few to be (he individual graves of every member of that setllcincnt. Probably because the precinct mound was rather low and tluis easily de­ stroyed in later centuries, the actual burials are rarely recovered. From the few well-preserved cases, however，we know that in eastern Honshu the precinct usually contained a single central pit or coffin burial with occasional extra interments in the surrounding inoaL Let us look at some actual examples of moated precincts. At Miya- nohara 宮ノ原in Kanagawa, a small Middle Yayoi settlement unit com­ prised one laree pit house, two to three small pit houses, and one square moated precinct. Such a site lends itself to interpretation as t he remains of an extended family where the lar(je building' was the house of the family head and the moated precinct his grave (Komiya 1988, p. 153; Nakayama 1989, p. 270).12 The type of unit apparently represented by M iyanohara conforms to whai K o n d o Yoshiro (1959) has termed a taii i shudan 単位集団一literally “unit eroup，” but which 1 translate here as

12 It is assumed that the family head '\,as，in fact, male, and Harunari (1991c, p . 101) sees this as evidence for patrilocal residence in eastern Japan at that time. H u d s o n ： An Archaeology of Yayoi Ritual 165

“corporate group.” Based on shared storage and paddy management, such groups have been widely posited for Yayoi society (e.g., T sude 1984; Nakayama 1988). The corporate group is a concept that is partic­ ularly effective for explaining moated precincts, though it assumes an unlikely equality between the component groups in any village. Otsuka 大塚，another site in Kanagawa, demonstrates this on a larger scale than Miyanohara. A moated village of about 22,000 m2，the Otsuka site dates primarily to the Miyanodai phase of the Middle Yayoi. Twenty-four square moated burial precincts were excavated 80 m southeast of the vil­ lage, with which they were contemporary.13 Based on his analysis of this site, Komiya (1975) proposed the following hypotheses：(1 )that each precinct was the grave of the smallest family unit, but (2) that there was considerable difference beLween Lhe person imerrecl in die center of the precinct and those buried in the moat; (3) that the cemetery could be divided into three main groups, which (4) could in turn be linked with three settlement clusters in the actual village (see also T sude 1984，pp. 136-39). These settlement and burial clusters seem to be the arcliaeo- loeical correlates of corporate groups, possibly based on agnatic descent. Even in east Honshu not all moated-precinct cemeteries follow ex­ actly the same pattern as Otsuka and Miyanohara. In the Kinai quite different conditions obtained, with much clearer evidence of social stratification. Certain Kinai moated prccincts were functionally more similar to the Kyushu mound-burials described earlier than to the Kanto precincts. At Kami カロ美 in Osaka the late Middle Yayoi moated mound no. Y-l had dimensions of 26 x 15m and was 2m lii

Die burial precincts were excavated separately and named the Saikacliido 歳 勝 土 site. 166 Japanese Journal o f Religious Studies 19/2-3 but remarkably standardized style of tomb mounds. These kofun will be discussed in the next section.

If preservation is good, it is usually easy to recognize prehistoric burial remains. A much harder task is to determine what sort of ceremonies were associated with the grave and its social and ideological meanings within a particular culture. In the case of the Yayoi we are fortunate that the Wei zhi provides some details of Wa funerals:

W hen a person dies, they prepare a single coffin, without an outer one. They cover the graves with [earth] to make a mound. W hen death occurs, m ourning is observed for more than ten days, dur­ ing which period they do not eat meat. The head mourners wail and lament, while friends sing, dance, and drink liquor. When the funeral is over, all members of the whole family go into the water to cleanse themselves in a bath of purification. This passage is often taken as evidence of temporary interment of the type known in early historic Japan.34 The obvious concern with purifi­ cation is reminiscent of later practices, though here it is strongly contrasted with a long period of uncleanliness during mourning. This contrast is developed in the next section dealing with the so-called “for­ tune keeper”：

W hen they go on voyages across the sea lo visit China, they always select a man who does not arrange his hair, does not rid himself of fleas, lets his clothing get as dirty as it will, does not eat meat, and does not approach women. This man behaves Like a mourner and is known as the fortune keeper. When the voyage turns out pro­ pitious, they all lavish on him slaves and other valuables. In case there is disease or mishap, they kill him, saying that he was not scrupulous in his duties, [emphasis added] The explicit analogy between the fortune keeper and the mourners sueeests pollution may have been seen as a liminal state with its own spiritual power. Some Yayoi mortuary rites are visible archaeologically. Features exca­ vated around the Yoshinogari mound, for instance, have led to various suggestions as to ritual activities that may have been associated with that burial facility. Red-burnished ceremonial pottery found along a path that led from outside the village to the south “entrance” of the mound may have been used in rites conducted for the dead. T akashim a (1990, p. 194) argues that the Yoshinogari mound served as a ritual focus of the

14 The text may be interpreted as refering to this practice, but my statement in HUDSON and Barnes(1991，p. 218) that structures for temporary interment are mentioned in the Wei zhi is incorrect. H u d s o n ： An Archaeology of Yayoi Ritual 167 local polity centered on the site, and that in this function it can be compared to later kofun mounds. The majority of moated burial precincts produce pottery with a hole intentionally knocked in the side or bottom of the vessel (fig. 9). Tashiro (1987, pp. 130-31) sug­ gests that, rather than containing offerings for the dead, these vessels were used in funeral ceremonies like those described in the Wei zhi, after which they were symbolically destroyed and abandoned in the moat to remove the pollution of death from the living. Tashiro's Fig. 9: Jar with hole knocked theory has been criticized by I t 6 into the side. Found in the (1988) and Y a m a g is h i (1989; 1991, moat of precinct No. 4 at Kane- p. 133) because it seems unable to no-O, Yamanashi. Height, 43 explain the wide variation in find cm. Late Yayoi. location and type of pottery. Some vessels, for example, are found lying on the bottom of the 丨noat, while others appear to have been put in place after a ccrtain amount of time had elapsed. Whatever the exact nature of lhe behavior represented by these vessels, it seems clear that wc are dealing with funerary rites of some sort, rites that continued and became more formalized in the Kofun era.

This section has provided a general overview of Yayoi-period m or­ tuary customs. Most of the discussion concentrated on classification of burial types and reconstruction of the social organization that the graves may represent. Both topics reflect the primary emphases of the Japan­ ese archaeological literature. The first may seem a perfectly neutral ex­ ercise, but even here we should be aware that we are forcing our own contemporary perspectives onto the past. In choosing whether to stress the mound or the moat of a grave, for example, we work on assumptions that may not have been held by the people who built thai grave. As men­ tioned in the introduction to this issue, there arc also many problems with reconstructing social organization from mortuary remains. Al­ though there has been little or no explicit discussion of the underlying assumptions, the approach to mortuary variability adopted by Japanese archaeologists is in many ways similar to that prevalent in Europe and America until the 1980s. Increasing burial complexity— as defined by the am oum of labor involved in construction and the size of the grave — is held to mirror increasing social stratification (cf. Tanigawa in this 168 Japanese Journal of Religious Studies 19/2-3 issue); development from communal to individual graves is another commonly posited scenario (e.g. Takakura 1973，1990). In the West this type of approach has been widely criticized over the past decade, mainly by archaeologists working within the British post-processual tradition. The main criticism is that, as B r a d l e y (1989，p. 448) puts it, the dead do not bury themselves. Burials are not simply the passive reflection of so­ cial relationships — they are actively created in the context of complex interests and social strategies. For such ideological reasons, real status may be masked or distorted in burials. Interestingly, there was a partial realization of this problem at an early date in Japan. As noted above, there is an obvious difference between the quantity of grave goods in Yayoi Kyushu and the Kinai, a difference that cannot be explained by social status alone since the latter region was equally or more politically advanced than the former. Thus there could be no direct link between grave goods and social stratification (Takakura 1973，p. 7). Despite this important exception, however，mortuary vari­ ability has continued to be interpreted in a very passive way in Japan. The post-processual critiques do not mean that the earlier approaches have no value whatsoever. Rather, like M o r r is (1991) we should attempt to combine the best of both worlds. Many aspects of Yayoi mortuary be­ havior can be explained by the “traditional” hypotheses of Sa x e (1970), B in f o r d (1971)，and others. With the rise of full-scale food production, the development of formal cemeteries can be linked to an increasing concern with ancestral rights to land. There is also a general correlation between mortuary heterogeneity, effort expenditure, and social stratification.15 What these hypotheses cannot explain, however, is the range of variability in Yayoi burial customs and clearly ideological fac­ tors like the scarcity of grave-goods in the Kinai. The ostentatious grave- goods displays of Yayoi Kyushu represent exactly the sort of behavior that is best understood by a symbolic approach (cf. C a n n o n 1989), and I look forward to further work on both of these aspects of Yayoi burials.

Power, Gender, and Ideology: A Third-Century Religious Revolution^

In the first section of this paper I discussed ritual change between the Jom on and the Yayoi. We must now move on to look at the transition from Yayoi to Kofun. The idea that this transition was marked by major ethnic, political, and religious changes has long been a common theme in Japanese scholarship.16 In a recent issue of this journal, Robert Ell-

Depending on one’s definition, forma] cemeteries may be said to have been present in the Jomon period to some extent. One example would be the kanjodori 環状土簾 of Late Hokkaido. Buria] heterogeneity, however, clearly increased in the Yayoi. 16 O f course there are numerous theories，which differ in matters of detail; in his H u d s o n ： An Archaeology of Yayoi Ritual 169 wood (1990) incorporated much of the traditional non-archaeological data relating to religious change in a theory that has a few original twists of its own. Ellwood lakes the Nihon shoki and Kojiki accounts of the Em­ peror Sujin's reign as the basis of his theory. This is a natural place to start, since all scholars are in agreement that important changes of some sort occurred during that reign. Sujin’s was Miinaki-iri- biko-iniwe-no-mikoto; the iri-biko element means “incoming princc’’’ and the Kojiki further relates that his reign “was called [that of] Emperor Mimaki who first ruled the land” (Philippi 1968，p. 208). In his “Horsc- riders” theory Eg ami (1964) linked Mimaki with in south Korea and turned Sujin into a Korean usurper; in contrast, B a r n e s has made a good case that Mimaki was a Nara person (1988，pp. 18-19; see also 1984). Whatever his origins were, there seems little doubt that lie marks a major political break. In Mizuno YiVs influential three-dynasiy scheme, Sujin’s reign sees the start of the Old Dynasty, also known as the Miwa court since it was based around M l Miwa in the souihcasL Nara Basin (Mizuno 1952; cf. Barnes 1988，p . 11). Based on the historical accounts, E l l w o o d (1990, pp. 204-205) pro­ poses four basic themes of Sujin spiriiualiLy: 1 Female shamanism is repeatedly discredited, as in the story of the shamancss YaiiiaLo-to-iiiomo-so-biiTie, who died by stabbing her­ self in the genitals with a chopsiick lo atone for shaming her divine husband. 2 The simultaneous worship of both female (Amatcrasu) and male (Yaiiiato-no-Okunidaina) gods in the palace was thought to be in­ auspicious, causing Amatcrasu to bc removed to a nearby village in the care of a priestess, and later to be taken to lse.

3 A vertical cosmology of ascensions lo heaven via mouniains became important. 4 Sujin’s revelatory dreams appear to replace the trance-mediuin- sliip of shamanesses, suggesting male spiritual supremacy.

Ellwood argues that these elements are the opposite of those that can be inferred for the time of the Empress Jingo, when religion centered on female shamanesses and a horizontal cosmology where spirits came across the sea; there was “no reference to the sun and little to heaven or to the masculine land ” （E llw o o d 1990, p. 207). In the traditional chronology, of course, Sujin preceded Jingo, but Ellwood presents a convincing case that their order should be reversed. Associating Jingo with I lie queen Pimiko f Mimiko] 卑弥呼 mentioned in the Wei, zhi, he

“Horsenclers” theory Eeami Namio, for example, areues for a split between the Yayoi/Early Kofun and tlie Late Kofun (Eg AM I 196^). The actual oppositions involved., however, are re­ markably similar in a majority of cases. 170 Japanese Journal of Religious Studies 19/2-3 argues for a Sujin “religious revolution” that roughly coincides with the Yayoi-Kofun transition. It is beyond my expertise or the scope of this article to evaluate all of the historical aspects of Ellwood’s theory. W hat I want to do, however, is consider the archaeological evidence for ritual change at the end of the Yayoi to see how it may fit with hypotheses such as Ellwood’s. There were two main aspects of this change: the abandonment of many Yayoi ritual practices and the establishment of a new ideological scheme cen­ tered around tomb mounds. We have seen that bronze bells and weap­ ons—the major ritual artifacts of the Yayoi —dropped out of use in the Late phase. As Ishino succinctly puts it in his paper in this issue of the JJRS, at the end of the Yayoi bronze ritual implements were “smashed, discarded, and buried.” Ishino provides a number of concrete examples of that phenomenon. He also points out that such a dramatic negation of traditional ritual must have involved a major change in actual beliefs. Although his ideas are not developed in detail here, Ishino argues that Yayoi beliefs were replaced by a ritual complex based on mirrors and tomb mounds. The Kofun period is widely defined as the period in which /eo/zm 古墳 (“tomb mounds”）were in use (Y o s h id a 1979, p. 399). The question th us becomes: What are kofun} While a precise definition is difficult if not im ­ possible, there are three main factors we need to consider: plan, size, and associated furnishings. In all cases the key word is standardization. (1 ) Shape: Kofun were constructed in a hierarchy of shapes, within which the keyhole plan was the most prestigious. The origin of the key­ hole shape is unclear, although various theories have been proposed (cf. Amakasu 1977, pp. 35-38; Tsude in press). The contrast between the rather amorphous shapes of Yayoi mounds and the standardized sym­ metry of the keyhole tombs seems hard to explain in the absence of con­ tinental prototypes, yet recent research suggests that the front section may have developed from the projections on Yayoi mounds. The ideo­ logical significance of the keyhole shape is unexplained, although it may reflect Chinese cosmological beliefs about the circle and the square as representations oi heaven and earth. The rear part of the kofun (i.e., the circular part in the case of keyhole mounds) served as the actual burial. The front square section was attached onto this and is thought to have been a platform for conducting ceremonies for the deceased. (2) Size: Although small kofun also exist, the larger ones were of mon­ umental proportions, comparable in size with the of E?vpt and the other great tombs of the ancient world. Apart from sheer massive­ ness, m any kofun were also built to strict proportions (see H ojo 1989). (3) Burial facilities and erave goods: Kofun possess distinctive burial furnishings. In the early part of the Kofun period these included stone- lined burial chambers due down into the top of the mound, very long H u d s o n ： An Archaeology of Yayoi Ritual 171 cedar-log coffins, and grave goods comprising mirrors, swords, beads, and iron weapons and agricultural ; around the mound were placed facing stones and special ritual ceramic jars and stands, which later evolved into ifitlw . All of these features first appeared together at the Hashihaka 箸墓 kofun in Sakurai City, Nara in the late third century (fig. 10). Some 280 m long, Hashihaka is many times bigger than even the largest Yayoi mound. After Hashihaka, standardized keyhole-shaped kofun spread quickly, first around the Inland Sea and then east to the southern T6hoku in the fourth century. Shapes, tombs, and grave goods were re­ markably similar within the hierarchy of kofunt suggesting a shared ideological system amongst the central and regional elites. Some ele­ ments of the kofun ritual system show Chinese influence: T sude (in press) mentions coffin orientation, three-stepped rear mound construction, and a fondness for vermillion. In the Kinai and the regions directly influenced by Yamato power, chieftains were often buried with their heads pointing north (see fig. 2 in T sude in press). This was something that was not seen in the Yayoi period, and that Tsude believes indicates Confucian influence. The other two regions thai made major contribu- tions to the kofun burial system were Ryusliu and Kibi (Okayama). Fac­ ing stones and the jars and stands that were ancestral to haniwa were derived from the Kibi area (K o n d o 1986; Kondo and H a r u n a r i 1967). We saw above how some Yayoi burials in Kyushu were very rich in grave goods, contrasting with the generally poor contents of Kinai graves. In

F ig .10: Hashihaka tomb mound, Sakurai. From SlIIKAISlll et al. 1984. 172 Japanese Joam al of Religions Studies 19/2-3 the Kofun period，the custom of burying the dead with sumptuous offerings spread to the Kinai. More importantly, the combination of mirror, sword, and curved jewel ( 勾玉) —which first appeared in Kyushu in the Middle Yayoi —was adopted by the Yamato polity as a symbol of its authority. As we defined kofun on the basis of standardization, it may seem strange to talk about “pre-standardized” kofun. Japanese archaeologists, however, use this term to refer to Late Yayoi mound burials that share certain pan-regional features. Terasawa (1988) has argued for the exis­ tence of what he calls “Makimuku-type keyhole tombs" dating to before Hashihaka. He defines this category primarily on the basis of plan pro­ portions where the diameter of the rear circular mound is twice the length of the front projection. Like the building at the Makimuku 纏向 site de­ scribed by Is h in o (this issue), these kofun seem to have been built to a 32 cm unit of measurement known as the Lu Ban 魯班 shaku. Originating in eastern Cnina during the Warring States era, this unit was considered auspicious and was used for temples and palaces in China; it is still used in contemporary Japan for temples and shrines, and is now known as the mon-jakn 門 尺 . Terasawa (1988, p. 105) provides a table of early kofun dimensions that appear very standardized when convened into this unit. The name “Makimuku-type” signifies his belief that these tombs drew their inspiration from the kofun of the Makimuku cluster in Sakurai, Nara, which was the center of the earliest Yamato polity. Mounds of this type, however, are found throughout a wide area from Shikoku to Chiba. The earliest, Tatetsuki 楣築 in Okayama，dates from the end of the Late Yayoi (cf. K ondo 1986). Tatetsuki is unusual in that it had two projections—both of which, however, fit Terasawa^ Makimuku- type dimensions. The first easily recognizable keyhole form is found at Makimuku Ishizuka 石塚 ，which probably dates to the early Shonai phase. There can be little doubt, therefore, that major ritual changes did occur at the end of the Yayoi period. Before we can relate these archae­ ological changes to historically-based theories such as Ellwood^ uSujin religious revolution,” however, we need to consider Lheir exact chronol- oev. Early pre-standardized kofun appeared in the Shonai ceramic phase or its regional equivalents. While there is considerable controversy over the time span occupied by this phase, most Japanese archaeologists place it within the third century. T erasa w a (1991b，p p . 117—18) argues that it began in the first and ended in the last quarter of that century. Terasawa’s dates are slightly earlier than others, but there seems lo be wide aereement that Shonai was produced during Pimiko’s reign (Tsude and Y am am oto 1990, pp. 154-55； S ah ara 1990, p. 186). Ii was still used after Pimiko’s death arouncl 250, and, since the Shonai ritual changes gained momentum as that phase proceeded, one could proba­ bly argue (based on our present chronological understanding，or lack of H u d s o n ： An Archaeology of Yayoi Ritual 173 it) that a “religious revolution” occurred after Pimiko. This would fit with Ellvvood’s theory, but in reality most Japanese scholars have sug­ gested that the major changes happened during the time of Pimiko. According to the Hou H an shu, the country of Wa was in a state of un­ rest during the reigns of Huan-di 桓帝（147-168) and Ling-di 霊帝（168- 189)’ although the Liang shu 梁書 narrows this down to the Guanglie 光和 era (178-85) of the latter reign. Both the Wei and Han histories imply that Pimiko was made queen by communal consent in order to resolve the political problems of this unrest. The choice of a female apparently went somewhat against tradition:

[] formerly had a man as ruler. For some seventy or eighty years after that there were disturbances and warfare. Thereupon the people agreed upon a woman for their ruler. Her name was Pimiko. She occupied herself with magic and sorcery, bewitching the people. Though mature in age, she remained unmarried. She had a younger brother who assisted her in ruling the country. After she became the ruler, there were few who saw her. She had one thousand women as attendants, but only one man. He served her food and drink and acted as a medium of communication.

This passage in the Wei zhi, more than any other, has fueled specula­ tion about ritual and politics in late Yayoi Japan. Almost all scholars take it to mean that Pimiko’s reign was politically and ritually different from what had gone before; many also believe that she marked a new stage of ritual unification (e.g., Yamao 1990，pp. 129-37). If, as is widely ar- eued by both historians (e.g., Yamao 1990) and archaeologists (e.g., S h ira ish i 1991)，this Kinai ritual confederacy (Yamatai?) was a direct predecessor of the early Yamato polity represented by standardized kofun, then Pimiko’s reign takes on an added importance.17 W hat was the exact nature of the changes that occurred during the time of Pimiko? The political status of the Kinai (Yamatai?-Yamato) con­ federacy (both at the pre-standardized and standardized kofun stages) re­ mains poorly understood. Certainly we cannot call it a state as Japanese historians once did. Barnes (1988) has pointed out that the existence of a shared ideological system symbolized by the kofun does not necessarily imply complete political unification from the very beginning of the Kofun

17 Some comments on my view of Yamatai may be appropriate here.【believe that in the Late Yayoi there were several regional chiefdoms of varying decrees of size and complexity. Two large chiefdoms, or confederacies of chiefdoms, appear to have been centered in north Kyushu and the Kinai, and one of these was probably Yamatai. Various lines of evidence sup­ port the candidacy of both these areas, and it is possible that the compiler of the Wei zhi con­ fused the two. Since the Wei zhi mentions places that we know were in Kyushu, a KyushCi location for Yamatai may be most reasonable. Such an identification, however, cannot avoid the fact that by the nud-tliird century, the Kinai polity was the more powerful of the two. 174 Japanese Journal of Religious Studies 19/2-3 period. Neither does it imply, however, that the spread of kofun was a mere fashion，or that it did not involve some degree of political subor­ dination by the provincial polities, possibly in return for access to iron. W ith respect to religious changes, our main source of evidence (apart from the archaeological data) is the Wei zhi passage quoted above. The “magic and sorcery” is Tsunoda and Goodrich’s translation of the Chi­ nese guidao 鬼 道 . Most scholars have seen this as some sort of shaman­ ism, although there is disagreement as to its onein and nature. Based on a comparative analysis of the use of the term m idao in the early Chi­ nese texts, U ed a (1976, pp. 79-80) argued it may have contained Daoist elements. Obayashi (1977, pp. 187-88) suggests guidao can be under­ stood as a type of “new religion” that suddenly eained popularity at a time of social crisis. Both O bayashi (19フフ，pp. 166-68) and I s h in o (this issue) take up the proposal by the climatic historian Y am a m o t o (e.g., 1980) that a Little Ice Age at the end of the 2nd century caused an economic crisis，and argue that the decline of traditional Yayoi ritual was due to its perceived failure in the face of this crisis. Another factor may have been the rejec­ tion of the traditional communal rituals based on bronzes after individ­ ual status became highly developed. Whatever the ultimate cause, the Chinese historical accounts provide a good case for major ritual and po­ litical changes occurring during the reien of Pimiko. Although the ar­ chaeological transition continued after Pim iko,1 believe that the major turning point occurred before her death. O f course，moving this “religious revolution” back into Pimiko’s time would seem to be fundamentally incompatible with Ellwood’s theory, since he sees her reign as the embodiment of the archaic, female, Yayoi order. At this point it would be easy to argue that the archaeological and historical records are by nature so different as to offer little hope of re­ solving a problem such as this. Their very lack of compatibility, however, means that the one cannot necessarily rule out the other. Three main points seem relevant here. First, when characterizing the social and religious nature of the Yayoi we must be careful to avoid stereotyped generalizations of the Yayoi/ ritual/matriarchical versus Kofun/political/patriarchical type. Ellwoocl, like many before him, falls into the easy trap of seeing the Yayoi as the structural opposite of the Kofun. Likewise, E g a m i (1964, pp. 48-52) con­ trasted a “peaceful, agricultural, magico-relieious, South-east Asian” Yayoi and Early Kofun period with a “practical, warlike ... North Asian” Late Kofun, yet I would argue that the latter description fits the Yayoi better than the former. O f course this is not to deny there may be some truth in such contrasts, but they should be viewed as theories to be tested rather than received knowledge. The second point is the need for careful and critical consideration of H u d s o n ： An Archaeology of Yayoi Ritual 175

the status of Pimiko. Ellw ood (1990，p. 211)argues for a “patriarchal revolution” that replaced the “female spirituality characteristic of the Jingo/Pimiko religious era. ...” As noted by Edwards (1985，p. 11〉，a similar view has been espoused by some Japanese scholars such as Sakima (1982[1926]) and Takamure (1954), who accept “the notion of matriarchy as formulated by Morgan and others, and [see] Japanese his­ tory as illustrating the shift from matriarchal to patriarchal society.” This 19th century idea of matriarchy as an evolutionary stage preceding patriarchy has now been discredited in the West (Ehrenberg 1989，pp. 63-66), and the ethnographic record “offers no basis for assuming that fem ale political roles were ever the rule ” (E d w a r d s 1985, p. 13). The work of historians and anthropologists who see Pimiko as a purely reli­ gious leader sharing power with her brother who held sccular authority (see Obayashi 1977, pp. 92-112) is also unconvincing, since the Wei, zhi specifically mentions Pimiko conducted diplomacy in her own name and ethnographic parallels provide “no model for interpreting shanianistic abilities as a basis for acmeving the kind of political power described for [Pimiko] in the [Wei zhi]yy (E d w a r d s 1985，p. 14). Edwards himself sees Pimiko not as filling the role of a female shamanistic queen, but as a woman who filled the primarily male role of a divine king. Finally, even if a transformation from a horizontal to a vertical cos­ mology and from female to male spiritual power did take place, such a change is likely to have taken a considerable time; the historical records may arguably telescope such a process into one reign, but the real change must have taken much longer. Archaeology is uniquely suited to the study of such long-term ritual phenomena, although it can be difficult to integrate archaeological data with that from other sources. All this may seem rather pessimistic, but it is not intended to be so. Ar­ chaeologically we can identify major ritual transformations that occurred in Japan during the 3rd century, a large chunk of which was occupied by the reien of Pimiko. The Chinese histories further imply that there were considerable religious and political changes during the time of this queen. To link these changes with those proposed by Ellwood and others from the native historical evidence is a difficult task, but one that should not bc shied away from. At the very least we can agree on the pivotal im ­ portance of the third century in Japanese religious history.

Conclusions: Ritual Continuity，Change, and the Yayoi

While writing this article I have been aware of two major methodologi­ cal problems that have no doubt been troubling many readers as well. These are the meaning of the ritual zones we recognized in the archae­ ological record and the question of how we measure and explain ritual continuity and change. The first problem is part of a wider debate within 176 Japanese Journal of Religious Studies 19/2-3 current archaeology as to the meaning of variability. Earlier scholars assumed a direct link between ethnicity and different types of pottery or stone tools. Recently archaeologists have stressed the many cultural factors that can interfere in such a relationship, but serious the­ oretical difficulties remain in their elucidation. This paper will have given some idea of the various ethnic, ideological, social, and political factors that can influence zone formation. Because of the phenomenal rate of excavation in Japan, the archaeological record there is suffi­ ciently fine-grained to make study of the social meaning of prehistoric cultural zones extremely rewarding, and it is hoped that this will soon become a focus of Japanese research. The second problem, of continuity versus change, also falls into a larger archaeological discourse as well as forming part of the debate on the origins of Japanese religion. In other words, we would like to know the relationship not just betweenjomon, Yayoi, and Kofun ritual behav­ ior, but also between Yayoi practices and “traditional” Japanese religion. Again, though, there are no easy answers. Apart from the question of recognizing and quantifying change in the archaeological record, there is the problem of assigning significance to those changes. In the case of the Yayoi, for example, we have seen that there was a distinctive set of burial customs and agricultural rites based around bronzes that were more or less unique to the Yayoi. On the other hand, we can recognize certain ritual practices that continue from the Jomon through the Yayoi and into modern Japanese society. One example is the use of miniature ceramic vessels in offerings. Similarities in such practices, however, do not rule out substantial changes in the meanings of the rituals con­ cerned: formal similarities between ritual “hardware” does not mean that the “software” has remained the same. It is for this reason that I am unconvinced by the work of Japanese scholars who see prehistoric rit­ ual not just as “primitive Shinto，” but as a pure form of a native cult that was gradually overlain by foreign influences (e.g., U m ehara 1992). As we have seen, the reality was quite difiFerenし Yayoi ritual was a rap­ idly-changing mixture of beliefs and behavior, further complicated by hieh regional variation. While it is lair to say the Yayoi saw quite intense continental influences, that does not mean Jomon ritual was a static phenomenon, any more “pure” than the other stages of Japanese reli­ gious history. While keeping these reservations in mind, however, we cannot ignore the possibility of long-term structures in Japanese ritual. The reasonably detailed picture of Yayoi ritual we now possess means that it is possible to begin serious study of those structures. Through archaeology，Japan’s ancient past can thus be interpreted in its own terms rather than through a priori assumptions derived from contemporary Japanese society. H u d s o n ： An Archaeology of Yayoi Ritual 177

REFERENCES

A makasu Ken 1977 The significance of the formation and distribution of kofun. Acta Asiatica 31:24—50. Aramaki Minoru荒巻実，W akasaT ^ru若狭徹，Miyazaki Shigeo宮崎重雄， Toyama Kazuo 外山和夫，and Iijima Yoshio 飯島義雄 1988 Oki II iseki ni okeru “sais6bo” no kozo: Shutsudo hone rui no bunseki kara 沖 II 遺跡における「再葬墓」の構造一出土骨類の 分析から[On the structure of secondary burial at the Oki II site, Gunma Prefecture, Central Japan, from the study of the human bones]. Gunma-ken ritsu rekishi hakubutsukan kiyd 9: 59­ 98 (with English summary).

Ba r n es ，Gina L. 1984 Mimaki and the matching game. Archaeological Review from Cambridge 3/2: 37-47. 1988 Protohistonc Yamato: Archaeology of the First Japanese State. Ann Arbor: University of Michigan Center for Japanese Studies and the Museum of Anthropology, University of Michigan. 1990 Early Korean states: A review of historical interpretation. In Hoabinhian, Jomon, Yayoi, Early Korean States: Bibliographic Re­ views of Far Eastern Archaeology 1990, Gina L. Barnes, ed.，113- 62. Oxford: Oxbow.

B in f o r d , Lewis R. 1971 Mortuary practices: Their study and their potential. In Ap­ proaches to the Social Dimensions of Moituayy Practices, J. A. Brown, ed., 6-29. Memoirs of the Society for American Archaeoloev 25.

Bla c k e r, Carm en 1981 Japan. In Divination and Oracles, Michael Loewe and Carmen Blacker, eds.，63-86，London: Allen and Unwin. 1983 Divination. Entry in Kodansha Encyclopedia of Japan. Tokyo: Kodansha.

B r a d l e y , Richard 1989 Comment on Aubrey Cannon, “The historical dimension in mortuary expressions of status and sentiment.” Current Anthro­ pology 30/4: 448-49.

Ca n n o n ，Aubrey 1989 The historical dimension in mortuary expressions of status and sentiment. Current Anthropology 30/4: 437-58. Doi Takashi 土肥孝 1984 Shuryo girei kara noko girei e 狩獵儀礼から農耕儀礼へ[From huntine to agricultural ceremonies]. In Jomon kara Yawi e 縄文 178 Japanese Journal of Religious Studies 19/2-3

から弥生へ[FromJOmon to Yayoi], pp. 147-57. Edited and pub­ lished by Tezukayama KOkogaku Kenkyujo, Nara.

Ed w a rd s, Walter 1985 Interpreting : Gender assumptions in the historiogra­ phy of a third-century Japanese queen. Unpublished ms.

E gami Nam io 1964 The formation of the people and the origin of the state in Japan. Memoirs o f the Toyo Bunko 23: 35-70.

Eh r e n b e r g , Margaret 1989 Women in . London: Publications.

E l l w o o d , Robert S. 1990 The Sujin religious revolution. Japanese Journal of Religious Studies 17/2-3: 199-217. Fujita Sabur6 藤田三郎 1986 Karako, Kagi iseki 唐 古 • 鍵 遺 驗 [The Karako-Kagi site] 22nd, 24th and 25th seasons. Nara: Tawaramoto-cho KyOiku Iinkai. 1988 Karako, Kagi iseki [The Karako-Kagi site] 21st and 23rd seasons. Nara: Tawaramoto-cho KyOiku Iinkai. 1989 Karako, Kagi iseki [The Karako-Kagi site] 32nd and 33rd sea­ sons. Nara: Tawaramoto-cho KyOiku Iinkai.

Fukuoka Shiritsu Rekishi Shiry6kan 福岡市立歴史資料館，eds. 1986 Sawara 6bo to sono jid a i 早良王墓とその時代[The Sawara royal tombs and the period in which they were built]. Fukuoka: Fukuoka Shiritsu Rekishi Shiryokan.

Harxjnari Hideji 春成秀爾 1982 Dotaku no jidai 銅鐸の時代[The age of ddtaku]. Bulletin of the N a­ tional Museum of Japanese History 1：1-48 (English summary). 1987a Dotaku no matsuri 銅鐸のまつり[Rituals involving dotaku bronze bells]. Bulletin of the National Museum ofJapanese History 12:1-37. (English summary) 1987b Basshi 抜歯[Tooth ablation]. In Yayoi bunka no kenkyu 8: M atsun to haka to yosooi, Kanaseki Sc Sahara, eds., 79-90. 1988 Maiso no shomondai 埋葬の諸問題[Various problems pertain­ ing to burials]. In Ikawazu iseki 伊川津遗跡[The Ikawazu Site], Onoda et al” eds., 395-420. Atsumi, Aichi: Atsumi-ch6 KyOiki Iinkai. 1989 Kyushu no dCtaku 九州の銅鋒[Ddtaku from Kyushu]. KOkogaku zasshi 75/2: 1-50. 1990a Otoko to onna no tatakai 男と女の闘い[The interpretation of a drawing of a fight sketched on ddtaku]. Bulletin of the National Museum of Japanese History 25:1-30 (English summary). H u d s o n ： An Archaeology of Yayoi Ritual 179

1990b Yayoi jidai no hajimari 弥生時代の始まり[The beginning of the Yayoi Period]. Tokyo: University of Tokyo Press. 1991a Tsuno no nai shika: Yayoi jidai no noko girei 角のない鹿一弥生 時代の農耕儀礼[Deer without antlers: Agricultural rituals of the Yayoi period]. In Nihon n i okem shoki Yayoi bunka no semtsu 日本 における初期弥生文化の成立[The formation of the earliest Yayoi culture in Japan], Takakura Hiroaki，ed., 442-81. Fuku­ oka: Prof. Yokoyama KGichi’s Retirement Memorial Committee. 1991b Kaiga kara kigo e: Yayoi jidai ni okeru noko girei no seisui 絵画から記号へ_ 弥生時代における農耕儀礼の盛衰[From pic­ tures to siens: The rise and fall of Yayoi am cultural ritual] B ul­ letin of the National Museum ofJapanese History 35:1-65 (English summary). 1991c Shinzoku soshiki no henka 親族組織の変化[Changes in family organization]. In Yayoi bunka: Nihon bunka no genyyu o saguru [Yayoi culture: In pursuit of the origins of Japanese culture], Museum of Yayoi Culture, eds., 97-102. Tokyo: Heibonsha.

H a s h ig u c h i, Naotake n.d. The lzu Islands and the prehistoric and ancient culture of eastern Japan. Ms submitted to Asian Perspectives.

H o d d e r , lan 1990 The Domestication o f Europe. Oxford: Basil Blackwell.

Hojo Yoshitaka 1989 The study of keyhole shaped tombs and Japanese archaeology. Archaeological Review from Cambyidge 8/1: 81-90.

Hoshida Ryoji 星田享ニ 1976 Hieashi Nihon Yayoi jidai shoto no doki to bosei東日本弥生時代 初頭の土器と墓制[Pottery and burials of the earliest Yayoi pe­ riod in east Japan]. Shtfian 7:10-52.

H u d s o n , M ark J. 1990 From Toro to Yoshinogari: Chaneine perspectives on Yayoi period archeology. In Hoabinhian, Jomon, Yayoi, Early Korean States: Bibliographic Reviews o f Far Eastern Archaeology 1990, Gina L. Barnes, ed., 63-111. Oxford: Oxbow. 1991 Miyanomae: An early rice paddy site in Yamanashi. EAANnouncements [Newsletter of the East Asian Archaeolop^ Network] 5:13-14.

H u d s o n , Mark J. and Gina L. Barnes 1991 Yoshinogari: A Yayoi settlement in northern Kyushu. Monu- menta Nipponica 46/2: 211-35. 180 Japanese Journal o f Religions Studies 19/2-3

Iijima Yoshio 飯島義雄，Miyazaki Shigeo 宮崎重雄，and Toyama Kazuo 外山和夫 1986 Yatsuhagi dokutsu iseki shutsudo jinkotsu ni okeru basshi no keifu八束脛洞窟遺跡出土人骨における抜歯の系譜[On the extrac­ tion oi human teeth in the Yavoi period from the Yatsuhagi cave]. Gunma-ken litsu rekishi hakulmtsukan kiyd 7: 45-74 (with English summary). 1987 Iwayuru usaisobon no saikent6 ni mukete no yosatsu—Tokuni shutsudo honerui ni shOten o atete所 謂 「再葬童」の再検討に向け ての予察一特に出土骨類に焦点をあてて[Preliminary report of the reinvestigation on so-called “secondary burial，’ 一with spe­ cial emphasis on the human bones]. Gunma-ken ritsu rekishi hakubutsukan kiyo 8: 21-50 (with English summary).

Inoue Y6ichi 井上洋ー 1990 lnoshishi kara shika c: D6butsu isho kara mita J 6111011 shakai

kara Yayoi shakai e 110 henkaイノシシからシカへ一 動物意匠から みた縄文社会から弥生社会への変化[From wild boar to deer: An aspect of change in prehistoric society seen in the artistic representation of animals]. Kokugakuin Daigaku kdkogaku shiryokan kiyd 6: 27-39.

Ishikawa Hideshi石川日出志 1981 Mikawa, Owari ni okeru Yayoi bunka no seiritsu 三河•尾張 における弥生文化の成立[The formation of Yayoi culture in the Mikawa and Owari regions]. Sundai shigaku 52. 1987 Saisobo 再葬墓[Secondary burials]. In Yayoi bunka no kenkyu 8: Matsuri to haka to yosooi, Kanaseki 8c S ahara, cd., 148-53. 1988 Jomon, Yayoi jidai no sh6jinkotsu 縄文、弥生時代の焼人骨 [Burnt hum an bones of the jOmon and Yayoi periods]. Sundai shigaku 74: 84-110. Ito Toshiyuki伊藤敏行 1988 T^kyo wan seigan ryuiki ni okeru hokeishukobo no kenkyu II 東京湾西岸流域における方形周溝墓の研究 ii [A study on the square moated burial precincts of the western drainage of Tokyo Bay II]. Tdky6-to maizO bunkazai sentd kenkyu ronshii 6: レ 69. (English summary)

Iwasaki Jiro 岩崎ニ郎 1987 Shisekibo 支石墓[ burials]. In Yayoi bunka no kenkyu 8: M atsuri to haka to yosooi, K anaseki 8c Sahara, eds., 91-97.

Kanaseki Hiroshi 金関恕 1986 Jujutsu to matsuri 祝術と祭[Magic and ritual]. In Iwanami koza Nihon kohogaku 4: ShUraku to saishi岩波講座日‘ 考 古 学 4 : 集落と 祭祀[The Iwanami Archaeology of Japan 4: Settlements and H u dso n : An Archaeology of Yayoi Ritual 181

rituals], Kondo Yoshiro et al., eds” 269-306. Tokyo: Iwanami Shoten.

Kanaseki Hiroshi and Sah ara Makoto, eds. 1987 Yayoi bunka no kenkyu 8: Matsuri to haka to yosooi 弥生文化の研究 8:祭と墓と装い[Research on Yayoi culture 8: Ritual, graves and ornaments]. Tokyo: Yuzankaku.

K a n e k o Erika 1968 A review of Yayoi burial practices. Asian Perspectives 9 (1966): 1-26. Kanzawa Yuichi 神澤勇ー 1990 Jujutsu no sekai: Kotsuboku no matsuri ES術の世界： 骨卜の まつり[A world of sorcery: Oracle bone rituals]. In Yayoijin to matsun 弥生人と祭[Rituals and the ], Ishikawa Hideshi, ed., 67-107. Tokyo: Rokko. Kashiwakura Ryokichi 柏倉亮吉 1961 Misakiyama shutsudo no seidoto 三崎山出土の青銅刀[A bronze knife excavated at Misakiyama]. Tohonu kokogaku 1:1-12. Kataoka Koji 片岡宏ニ 1991 Kyushu chiho no Yayoi-bo 九州地方の弥生墓[Yayoi graves in Kyushu]. In Genshi，kodai Nihon no bosei 原子、古代日本の墓制 [The burial systems of prehistoric and ancient Japan], Yama­ gishi Ryoji, ed” 188-204. Tokyo: Dosei.

K id d e r , J. Edward, Jr. 1991 Ritual in the late Jom on period. Paper presented at the 12th Congress of the International Union of Prehistoric and Pro to- historic Sciences, Bratislava, Czechoslovakia，September. Kobayashi Hirokazu and Satomura Koichi 里村晃ー 1987 Mura no rekishi—Kodai 村の歴史一古代[The history of the vil­ lage: Ancient period]. In Mukawa sonshi [A history of Mukawa village], 235-65. Kobayashi Tatsuo 小林達雄 1991 Jomon bunka to Yayoi bunka 縄文文イ匕と弥生文{匕[jGmon cul­ ture and Yayoi culture]. In Yawi bunka: Nihon bunka no p-enryu o sagum y Museum of Yayoi Culture, eds., 212-17. 1992 Regional organization in Jomon Japan. Translated with Intro­ duction by Mark Hudson and Mariko Yamagata. Arctic Anthro­ pology 29/1. Kobayashi Yukio 小林行雄 1959 Kofun no hanashi 古 墳 の 話 [Talking about kofun]. Tokyo: Iwanami. Kom iya T suneo小宮恒夫 1975 Yayoi jidai chuki no hokeishukobo 弥生時代中期の方形周溝墓 182 Japanese Journal of Religious Studies 19/2-3

[Middle Yayoi square moated burial precincts]. In Saikachido iseki [The Saikachido site], Sakaue Katsuhiro and Sakamoto Akira, eds., 159-84. Yokohama: Yokohama-shi Maizo Bunkazai Chosa Iinkai. 1988 Kanagawa-ken tobu no hokeishukobo神奈川県東部の方形周溝墓 [Square moated burial precincts of east Kanagawa]. In Higashi Nihon no Yayoi bosei 東日本の弥生墓制[The Burial Systems of Yayoi East Japan]，Kita Musashi Kodai Bunka Kenkyukai, eds., 140-70. Gunma: Gunma-ken Kokogaku Kenkyujo, Chikuma- gawa Suikei Kodai Bunka Kenkyujo, and Kita Musashi Kodai Bunka Kenkyukai.

Kondo Takaichi 近藤喬ー 1987a Seidosei saigu wa naze umeraretaka 青銅製祭具はなぜ埋め られた；^ [Why were bronze ritual implements buried?]. Round table discussion with Ishino Hironobu in Kofun hassei zengo no kodai N ihon 古墳発生前後の古代日本[Ancient Japan at the time of the appearance of tomb mounds], Ishino Hironobu, ed., 107-132. Tokyo: Owa Shobo. 1987b Seidoki to tekki: Juyo to tenkai青銅器と鉄器：受容と展開[Bronze and iron implements: Acceptance and development]. In Asahi hyakka Nihon no rekishi [The Asahi encyclopedia of Japanese his­ tory] 39:172-78. 1987c Seidoki o megutte 青銅 器を めぐっ て [An overview of bronze implements]. In Kofun hassei zengo no kodai Nihon, Ishino Hiro­ nobu, ed., 90-106..

Kondo Yoshiro 近藤義郎 1959 Kyodotai to tan’i shudan 共同体と単位集団[Corporate groups and the community]. Kokogaku kenkyu 6/1: 13—20. 1986 The keyhole and its relationship to earlier forms of burial. In Windows on the Japanese Past: Studies in Archaeology and Prehistoyy，R. Pearson, G_ Barnes, and K. Hutterer, eds., 335­ 48. Ann Arbor: University of Michigan Center for Japanese Studies.

K o n d o Yoshiro and H a r u n a r i Hideji 1967 Haniw a no kigen 埴輪の起源[The origins of haniwa]. Kokogaku kenkyu 13/3:13-35.

Kuraku Yoshiyuki エ楽善通 1990 Dotaku no matsuri 銅驛のまつり[Ddtaku rituals]. In Yayoijin to mutsun [Rituals and the Yayoi people], Ishikawa Hideshi, ed., 29-65. Tokyo: Rokko. H u d s o n ： An Archaeology of Yayoi Ritual 183

Mishina Shoei三品彰英 1968 DOtaku shOkO 銅驛小考[Some thoughts on ddtaku). Chdsen gakuhd 49: 361-73. Miyashita 宮下健司 1983 ]6mon dogu no shuen 縄文土偶の終焉[The end of Jom on clay figurines]. Shinano 35/S: 78-101. Miyazaki Shigeo 宮崎重雄，Toyama Kazuo 外山和夫，and Iijima Yoshio 1985 Nihon scnsni jidai ni okeru hito no hone oyobi ha no senko ni t s u i t e 日本先史時代におけるヒトの骨および歯の穿孔について [Perforated human bones and teeth in prehistoric Japan]. Gunma-ken ritsu rekism hakubutsukan kiyo 6: 77-108. M izuno Masayoshi 水野正好 1990 no dojO —Shimaguni no genzO文明の土壌一島国の原像 [The foundation of civilization —The original state of an island nation]. Tokyo: Kadokawa. M iz u n o Y Q 水野祐 1952 Zotei Nihon kodai ochO shiron josetsu 増訂日本古代王朝試論序説 [Revised introduction to theories of the ancient Japanese dy­ nasties). Toky6: Komiyama Shoten.

M o r r is , Ian 1991 The archaeology of ancestors: The Saxe/Goldstein hypothesis revisited. Cambridge Archaeological JourTial 1/2:14フー69. Nakahashi Takahiro 中橋孝博 and Nagai Masafumi 永井昌文 1987 Fukuoka-ken Shima-cho Shinmachi iseki shutsudo no Jomon, Yayoi ikoki no k o ts u 福岡県志摩町新町遺跡出土の縄文•弥生 移行期の人骨[Transidonal jomon-Yayoi human skeletal re­ mains from the Shinmachi site, Shima-chO, Fukuoka Prefec­ ture]. In Shinmachi iseki [The Shinmachi site], Hashiguchi Tatsuya, ed., 87-105. Shima, Fukuoka: Shima Municipal Board of Education. Nakamura Tomohiro 中村友博 1987 Buki-gata saiki 武器形祭器[Weapon-shaped ritual implements]. In Yayoi bunka no kenkyu 8: Matsuri to haka to yosooiy Kanaseki et al” eds” 25-31. Nakamura Yoshiyuki 中村良幸 1989 Nemuro-shi Hattaushi 20 iseki shutsudo no dogu 根室市初田牛 20 遗 跡 出 土 の 土 偶 [Clay figurines excavated from the Hattaushi 20 site, Nemuro]. In Hattaushi 20 iseki fiakkutsu chosa hokokuslio初田牛 2 0 遺跡発掘調査報告書[Excavation report of the Hattaushi 20 site], Kawakami Jun 川上淳 and Toyohara Shoji 豊原熙司，eds.，43-55. Nemuro, Hokkaido: Nemuro Board of Education. 184 Japanese Journal of Religious Studies 19/2-3

Nakayama Seiji 中山誠ニ 1988 Yayoi jidai shumatsu ni okeru Uenodaira iseki no shuraku kozo 弥生時代終末における上の平遺跡の集落構造[The settlement structure of the nnal Yayoi Uenodaira site]. Yamanashi Kenntsu Koko Hakubutsukan and Yamanashi-ken Maizobunkazai Senta Kenkyu kiyo 4: 31-51. 1989 Chubu, Kanto chiho no Yayoi shudanbo no kosei ni tsuite 中部， 関東地方の弥生集団墓の構成について[The structure of Yayoi community graves in the Chubu ana Kanto regions]. In Yamanashi kokogaku ronshu 山梨考古学論集[Papers on Yamanashi archaeology], Volume I I ，265-83. Niitsu Takeshi 新津健 1986 Yamanashi ni okeru Jomon bunka no dento to shometsu 山梨に おける縄文文化の伝統と消滅[j6mon culture in Yamanashi: Tra­ dition and decline]. In Yamanashi kokogaku ronshu, Volume I, 179-204. N iitsu Takeshi and Yamaki Yoshio 八卷與志夫，eds. 1989 咖 77 金生遺跡[The Kinsei site II]. Kofu: Yamanashi-ken Kyoiku Iinkai. Nishimoto Toyohiro 西本豊弘 1989 Shimogori-kuwanae iseki shutsudo no dobutsu itai 下郡桑苗遺跡 出土の動物遺体[Faunal remains excavated from the Shimogori- kuwanae site]. In Smmogori-kuwanae iseki [The Shimogori- kuwanae site], Takahashi Nobutake, ed., 48-61. Oita, Kyushu: Oita Prefectural Board of Education. 1991a Jom on jidai no shika, inoshishi shuryo 縄文時代のシカ， イノ シ 'ン狩撖[Deer and wild boar hunting in the Jomon period]. Kodai 9 1 :114-32. 1991b Yayoi jidai no buta ni tsuite 弥生時代のブタについて[Pigs in the Yayoi period]. Bulletin of the National Museum ofJapanese History 35:1-65.

Nomtjra Takashi 村野崇 1985 Hokkaido Jomon J id a i shumakki no kenkyu 北海道縄文時代終末期 の研究[A Study of the terminal Jomon Period in Hokkaido]. Sapporo: Miyama Shobo. Obayashi Taryo 大林太良 1977 Yamataikoku 邪馬台S . Tokyo: Chuo Koronsha. 1987 Jomon to Yayoi no haka: Minzokugaku teki kaishaku 縄文と 弥生の墓一民▲学 的 解 釈 [j6mon and Yayoi graves: An ethno­ graphic explanation]. In Yayoi bunka no kenkyu 8: M atsuri to haka to yosooi, Kanaseki 8c S ah ara.，eds., 165-73. H u d s o n ： An Archaeology of Yayoi Ritual 185

O d a F u jio 小田富士雄 1990 Yayoi jidai no hokubu Kyushu 弥生時代の北部九州[North Kyushu in the Yayoi period]. In Wakoku tairan to Yoshinogari 倭国大乱と吉野ケ里[Yoshinogari and the Wa Unrest], 71-115. Tokyo: Yamakawa. Okita Masaaki置田雅昭 in press Dotaku no hakken o motomete 銅鐸の発見をもとめて[On the trail of bronze bell nnds]. Tenn Daxmku omsato kenamjo kaiho 天理大学おやさと研究所会報6. Oshima Naoyuki 大島直行 1990 Kita no b u n k a :)匕の文イ匕[The culture of the north]. In Ningen no bijutsu 2: Ine to kenyyoku 人間の美術 2 : 稲と権力[Human art 2: Rice and power], Umehara Takeshi, general editor, 80-87. Tokyo: Gakushu Kenkyusha.

P h il ip p i, D on ald L. 1968 Kojiki. Tokyo: University of Tokyo Press.

P ig g o t t , Jo a n R. 1989 Sacral kingship and confederacy in early Izumo. Monumenta Nipponica 44/1:45-74. Saga Board of Education 佐賀県教育委員会，eds. 1990 Yoshinogari iseki gaiho 吉野ケ里遺赫概報[The Yoshinogari site: Preliminary report]. Tokyo: Yoshikawa Kobunkan. Sahara Makoto 佐原眞 1960 Dotaku no chQzo 銅鐸の鋳造[The casting of bronze bells]. In Sekai kokogaku taikei 世界考古学大系[World archaeology series] V ol.2，pp. 92-104. Tokyo: Heibonsha. 1980 Yayoi doki no kaiga 弥生土器の絵画[Pictures on Yayoi pottery]. Kokogaku zasshi 66/1: 102-17. 1987a The Yayoi culture. In Recent Archaeological Discoveries in Jap an , Tsuboi Kiyotari, ed., 37-54. Paris and Tokyo: UNESCO and The Centre for East Asian Cultural Studies. 1987b Nihonjin no tanjo [The birth of the Japanese]. Tokyo: Shogakkan. 1990 Round-table discussion in Wakoku tairan to Yoshinogari [Yoshinogari and the Wa Unrest], Otsuka Hatsushige et al” 117-247. Tokyo: Yamakawa. in press Rice growing and the Japanese. Acta Asiatica 63.

Sa h l in s , Marshall 1972 Economics. Chicago: Aldine. Sakima Koei佐喜真興英 1982 Nyomn seiji ko: J in r u i genshi kihon no kenkyu 女人政治考： 人類原始基本の研究[A consideration of female government: 186 Japanese Journal of Religious Studies 19/2-3

Basic research on primitive people]. Shisensha (reprint: 1st edi­ tion 1926).

Sa x e , A. A. 1970 Social dimensions of mortuary practices. Ph.D. thesis, Univer­ sity of Michigan. Published by University Microfilms, Ann Arbor, 1973. Shiraishi Taichiro 白石太一郎 1991 Yamataikoku jidai no Kinai, Tokai, Kanto 邪馬台国時代の幾内， 東海，関東[The Kinai, Tokai and Kanto in the age of Yamatai]. In Yamataikoku jidai no higashi Nihon 邪馬台国時代の東日本[East Japan in the age of Yamatai], National Museum of Japanese History, ed., 45-71. Tokyo: Rokko. Shiraishi Taichiro, Harunari Hideji 春成秀爾，Sugiyama Shinsaku 杉山晋作，and Oku da Hisashi奥田尚 1984 Hashihaka kofun no saikento 箸墓古墳の再検討[Rethinking the Hashihaka tomb]. Bulletin of the National Museum ofJapanese History 3: 41-80 (English summary pp. 276—77). Shitara Hiromi 設楽博巳 1988 Saisobo kenkyu no genjo to kadai再葬墓研究の現状と課題[Prob- lems on the study of secondary burials]. Kdkomku zasshi 74/2: 108-109. 1991a Yayoi jidai no noko girei 弥生時代の農耕儀礼[Agricultural rit­ uals of the Yayoi period]. Kikan kokogaku 37: 59-64. 1991b Kanto chiho no Yayoi bunka 関東地方の弥生文化[The Yayoi culture oi the Kanto]. In Yawi bunka: iNinon bunka no gemyu o sagwru、Museum of Yayoi Culture, eds.,196-202. Sugihara Sosuke 杉原荘介 1968 Fukushima-ken Narita ni okeru shotateana to shutsudo doki 福島県成田における小S 穴と出土土器[Smalls pits and their ex­ cavated pottery from Narita, ]. Kdkomku shukan 4/2: 19-28. 1981 Tochigi-ken Izuruhara ni okeru Yayoi jidai no saisoboeun 栃木県出流原における弥生時代の再葬墓群[The Yayoi period sec­ ondary burial cemetery at Izuruhara, Tochigi]. Meiii Univer­ sity, Faculty of Arts Research Report, Archaeology 8. Tokyo: Meiji University. Sugihara Sosuke and Otsuka Hatsushige 大塚初重 1974 Chiba-ken Tenjinmae ni okeru Yayoi jidai chuki no boshi 千葉県天神前における弥生時代中期の墓址[Cemeteries of Middle Yayoi aee at Tenjinmae, ]. Meiji University, Faculty of Arts Research Report, Archaeoloev 4. Tokyo: Meiji University. H u d s o n ： An Archaeology of Yayoi Ritual 187

Suzuki Hisashi 鈴木尚 1983 Hone kara mita Nihonjin no rutsu 骨から見た日本人のルーツ [The roots of the Japanese as seen from bones]. Tokyo: Iwanami Shinsho. Takakura Hiroaki 高倉洋彰 1973 Funbo kara mita Yayoi jidai shakai no hatten 墳墓からみた 弥生時代社会の1発展過程[The development process of Yayoi so­ ciety as seen from graves]. Kdkomku kenkyu 20/2:7-24. 1990 Haka ni arawareta Yayoi shakai no tokushitsu wa nani ka 墓にあ らわれた弥生社会の特會は何か[What distinguishing character­ istics of Yayoi societv are visible in graves?]. In Soten: Nihon no rekishi 争 点 ：日本の歴史[Points of debate: The ] V o l.1，Suzuki Kimio，ed., 255-70. Tokyo： Shinjinbutsu Juraisha. Takamure Itsue 高群逸枝 1954 Josei no rekishi 女性の歴史[A history of women]. T6ky6: Kodansha. Takashima Chuhei 高島忠平 1990 Comment in Snmpojumu: Wakoku Tairan to Yamataikoku [Round table discussion: The Wa Unrest and Yamatai]. In Wakoku tairan to Yoshinogari [Yoshinogari and the Wa Unrest],117-247. Tanaka Migaku 田中琢 1970 “Matsuri” kara “matsurigoto” e 「まつり」か ら 「まつりごと」へ [From ritual to ritual government]. In Kodai no Nihon 5: Kinki [Ancient Japan 5: the Kinki district], Tsuboi Kiyotari and Kishi Tosnio，eds., 44-59. Tokyo: Kadokawa Shoten. 1991 soran 倭人争乱[The Wa Unrest]. Tokyo: Shueisha. Tashiro Katsumi 田代克巳 1987 Hokeishukobo 方形周溝墓[Square moated burial precincts]. In Yayoi bunka no kenkyu 8: Matsuri to haka to yosooi, Kanaseki Sc Sa­ h ara, eds” 148-53. Terasawa Kaoru 寺沢薰 1988 Makimuku-gata zenpo-koenfun no chikuzo 纏向型前方後円墳 の 築 造 [The construction of Makimuku-type keyhole-shaped tombs]. In Kokogaku to gijutsu 考古学と技術[Archaeology and tech­ nology], Mori Koichi, ed., 99-111.: Doshisha University. 1989 Seidoki maino no igi 青銅器埋納の意義 [The significance of bronze hoards]. Kikan kdkomku 27: 66-74. 1991a Yayoi jidai no seidoki to sono matsuri 弥生時代の青銅器と そ の マ ツ リ [Yayoi-period bronzes and ritual]. In Kdkomku: Sono mikata to kaishaku 考 古 学 ：その見方と解釈[Archaeology: Ap­ proaches and interpretations]. V ol.1 ,Mori Koichi, ed., 139-84. Tokyo: Chikuma Shobo. 188 Japanese Journal of Religious Studies 19/2-3

1991b Yayoi jidai kara Kofun jidai e 弥生時代から古墳時代へ[From the Yayoi to the Kofun period]. Round table discussion with Owa Iwao. Higashi Ajia no kodai bunka 69:116-46.

T il l o t s o n , Dianne 1989 Mortuary patterning and the evolution of the rice ancestors. Indo-Pacific Prehistory Association Bulletin 9:1-14.

T oyam a Kazuo, M iyazaki Shigeo, and I ijim a Yoshio 1989 Saisobo ni okeru senko jinshikotsu no imi 最葬墓における 穿孔人歯骨の意味[The significance of perforated human teeth and bones in secondary burials], Gunma-ken ritsu rekishi hakubutsukan kiyd 10:1-30.

Tsude Hiroshi都出比呂志 1984 Noko shakai no keisei 農耕社会の形成[The formation of agri­ cultural society]. In Koza Nihon rekishi I: Genshi, kodai 1 [The his­ tory of Japan I: Prehistory and 1],Rekishigaku Kenkyukai and Nihonshi Kenkyukai, eds., 117-58. Tokyo: University of Tokyo Press. in press The K ofun period and state formation. Acta Asiatica 63.

Tsude Hiroshi and Yamamoto Saburo 山本三良P ，eds. 1990 Yamataikoku no iidai 邪 馬 台 国 の 時 代 [The age of Yamatai]. Tokyo: Mokujisha.

Tsunoda, R. and し C. Goodrich 1951 Ja p a n in the Chinese Dynastic Histories. South Pasadena, Ca.: PD and lone Perkins.

Ueda Masaaki上田正昭 1976 Wakoku no sekai 倭S の 世 界 [The world of the Wa]. Tokyo: Kodansha. 1986 A fresh look at ancient history.yapaw Quarterly 23/4: 403-409.

U m e hara Takeshi 梅原猛 1992 Nihonjin no tamashii 日本人の魂[The soul of the Japanese]. Tokyo: Kobunsha.

Watanabe Makoto 渡辺 _ 1991 Kunkok-ri kaizuka shutsudo no bokkotsu no kenkyfl郡谷里貝塚 出土のト骨の研究[A study of the oracle bones excavated from the Kunkok-ri shell ]. Nagoya Daigaku Bungakubu kenkyu ronshu (shigaku) 37:127-59.

W a ts u ji Tetsuro 和辻哲郎 1977 Nihon kodai bunka 日本古代文化[AncientJapanese culture]. In Watsuji Tetsuro zenshu 和辻哲郎全集[Collected works of Watsuji Tetsuro], v o l.3. Tokyo: Iwanami Shoten. H u d s o n ： An Archaeology of Yayoi Ritual 189

Yamagishi Ryoji 山岸良ニ 1989 Senko dokiron sobyo 穿孔土器論素描[An outline of theories on perforated ccramics]. Shikan 21:109-22. 1991 Saisobo 再葬墓[Secondary burials]. In Genshi, kodai Nihon no bosei 原始 • 古代日本の墓制[The burial systems of prehistoric and ancient Japan], Yamagishi Ryoji, ed., 105-19. Tokyo: Dosei. Yamamoto Takeo 山本武夫 1980 Ni, sail seiki no kike 2 • 3 世紀の気候[The climate of the 2nd and 3rd centuries]. In Sanseiki no kokogaku 三世紀の考古学[The archaeology of the third century]. Tokyo: Gakuseisha. Yamao Yukihisa 山尾幸久 1990 Sanseiki no Nihon 三世紀の日本[Japan in the third century). In Yamataikoku no jidai, T sude 8c Yamamoto, eds.,119-39. Yamauchi Noritsugu 山内紀_ 1987 Yosumi tosshutsubo四K 突出墓[Four-cornered burial mounds]. In Yayoi bunka no kenkyu 8: Matsuri to haka to yosooit K anaseki Sc Sa h a r a , eds., 142-47. Y o s h id a Shoichiro 吉田章一郎 1979 Kofun jidai 古 墳 時 代 [The Kofun Period], Entry in Sekai kokogaku jiten 世界考古学事典[An encyclopijedia of world archae­ ology]. Tokyo: Heibonsha.